Sunteți pe pagina 1din 98

[Type

a 0

De La Salle University – Dasmariñas

FORECASTING THE PRICE OF


CORN IN THE PHILIPPINES

A Thesis
Presented to the Faculty of the
Allied Business Department
College of Business Administration and Accountancy
De La Salle University-Dasmariñas
Dasmariñas City, Cavite

In partial fulfillment
of the requirements for the degree of
Bachelor of Science in Business Administration
(Major in Economics)

BILLY JULIUS M. GESTIADA

May 2018
1

De La Salle University – Dasmariñas

CHAPTER I

INTRODUCTION

According to the Agricultural Management Assistance (AMA) (2008) and the

Food and Agriculture Organization Corporate Statistical Database (FAOSTAT) (2012),

corn, or Zea mays or maize, belongs to the grass family, which originated in Central

America and belongs to the top three most grown cereal crops all over the world,

together with rice (Oryza sativa) and wheat (Triticum spp.). Its global commercial

production in 2010 reached 844.4 million metric tons, at which the harvested land is

161.9 million hectares.

Corn is the second most bountiful crop grown all over the world, and many

people have been consuming this for everyday living. It is a multifaceted crop, and

there is no wasted part on its plant. According to Gwirtz and Casal (2014), the two

basic categories applied in converting maize into other goods for human consumption

are dry and wet milling. In the wet milling process, maize is separated into the classes

of starch, protein, oil, and fiber. Once the separation has been done, the products are not

sold outright instead further industrial processing is required. After the said industrial

processes, the four chemical classes can now be sold as sweeteners that are either solid

or liquid. The dry milling process, on the other hand, involves particle size reduction of

maize, which maintains some of the maize germ and fiber. Again, the maize cannot be

sold right away for human consumption, not until some ingredients are further added

and some thermal processes such as boiling, drying, frying, and baking are applied.
2

De La Salle University – Dasmariñas

This will enable the finished product, though having its nutritional attributes altered

changed, to be sold for human consumption. Even so, there are various ways of

processing corn across various countries. After all, there are so many finished products

that can be extracted from corn more than anyone can imagine. In fact, Sailer (2012)

stated that corn husks in Mexico are made into their traditional tamale. Kernels are

converted into food. Animals feed on the stalks, and the corn silks are made into herbal

teas. Some food products like corn oil, corn meal, corn sweetener, corn syrup, and even

corn whiskey are made from corn.

In the United States (US), even if the farmers are capable of growing different

kinds of grains and crops and bringing them to the market, corn accounts for 90 percent

of all the produced grain. In 2015, about 80 million acres of farmland are being planted

with corn, and the world is being supplied with 20 percent of the American corn. While

it is true that the US is maintaining its current reputation as an international exporter of

corn, what remains from these corns is not entirely wasted. Given that corn is the

primary crop grown in the US, every man, woman, and child consumes four pounds of

corn a day, which amounts to a total of more than 1,500 pounds of corn consumed

annually. (NathanF, 2015)

Even though the US is considered to be the largest exporter of corn in the world,

less than 15 percent share of the demand for the US corn is accounted by the exports,

which is actually small. This occurrence has something to do with the demand-and-

supply-relationship of corn, resulting to the other markets adjusting to the US market’s

current price. Because of this internationally tough competition, farmers plant their corn
3

De La Salle University – Dasmariñas

after considering the size of the US crop in order to have a market advantage over the

short US crops. In fact, some countries like Brazil, India, and South Africa had

significant corn exports when international prices are competitive, or the crops are

large. (United States Department of Agriculture Economic Research Service [USDA

ERS], 2017)

In many countries, particularly the developing ones, commodities still remain a

reliable source of export earnings. Moreover, price movements of these commodities

play a major role on overall macroeconomic performance. Commodity-price forecasts

are essential in formulating and planning macroeconomic policies. (Bowman and

Husain, 2004)

These studies mentioned above are only a very small portion of numerous

studies done on commodity prices. In the field of economics, this kind of study is not

something new. The efforts of the previous researchers contributed a lot to the present

knowledge of commodity prices.

Background of the Study

The importance of forecasting commodity prices, corn included, remains a

prevailing issue at present. It is important to take a look at some of the techniques and

methods used by some researchers in forcasting corn prices, and how effective and

accurate these techniques are. A study was conducted by Halonen (2016) showing that

there are few statistical techniques that can outperform models that pertain to supply

and demand analysis in forecasting the price of corn in the US. The researcher argued
4

De La Salle University – Dasmariñas

that there are some econometric techniques that are costly to use, none of them of being

more costly than the supply and demand analysis. The main reason for this much

expense is that supply and demand analysis involves gathering and summarizing a large

amount of information regarding supply and demand. Furthermore, it also requires

extensive surveys to be distributed to a large sample in a particular study. That being

the case, this study examined if there are some statistical methodologies that can

provide forecasts at least as accurate, or even not as costly as the models incorporating

supply and demand analysis. Both the statistical methodologies and the supply and

demand models were evaluated at one, three, six, nine, and twelve month horizons,

given that these horizons are suitable for analyzing commodities that involve buying,

selling, production, and contract negotiations. It was found out that an AR model is the

best model to use in forecasting over a short horizon, while VAR model is the best

model to use in forecasting over a long horizon, over six months.

Another study pertaining to forecasting the price of corn, along with other 14

commodities, has been conducted by Bowman and Husain (2004). The research

analysed the performances of three different types of commodity price forecasts

namely: judgment-based, historical price-based, and commodity futures-based. Since

spot prices tend to move forward future prices for most commodities in the long run,

and the future prices showing lower variability, it was found out that commodity

futures-based model outperforms both the judgment- and historical price-based models

in directional terms, at the very least.


5

De La Salle University – Dasmariñas

Aside from the mentioned three different types of commodity price forecasts

above, Jha and Sinha (2013) conducted price forecasting on soybean and rapeseed-

mustard wholesale prices in India using neural network model. The researchers stated

that the innovation of Artificial Neural Network (ANN) proved to be feasible given the

data provided by developing countries. In this study, ANN indicated more significant

number of future price changes as compared to linear model. This means that in the

context of commodity price forecasting, where turning points are crucial, ANN model

might be preferred because it totally outperforms nonlinear models most especially

when the series is linear. Lastly, even if the series is nonlinear, combining linear and

nonlinear models was observed to perform better than these two models performing

independently.

From the studies mentioned above, it can be observed that there are a lot of

methods and techniques done in commodity price forecasting. The important issues are

the effectiveness and accuracy of these techniques, which are greatly changing over

time. This is the exact reason why the researchers never settle on existing forecasting

models, instead they either formulate their own models or they improve the existing

models that will be enough to account for the present, changing factors.

Although there have been many papers done in other countries pertaining to

models used in forecasting the price of corn, not much is done in the Philippines. This

paper will focus on providing an econometric model in forecasting the price of corn in

the Philippines.
6

De La Salle University – Dasmariñas

Statement of the Problem

Commodity price forecasting is an essential part of any industry involving

trading and price analysis. Commodity prices are often unpredictable that becomes

even highly unpredictable when you factor the presence of natural calamities droughts,

typhoons, floods, and pests. Because of this, there’s a greater risk and uncertainty in

formulating a forecasting methodology. In the case of the Philippines, where rice and

corn are the major crops, policy makers should see to it that they make reliable, highly

accurate forecasts of rice and corn prices in order to ensure food security, thus

somehow alleviating hunger and poverty. Farmers will also benefit from commodity

price forecasting because they will definitely want to make their production and

marketing decisions wisely so that they will be able reap positive financial outcomes in

the future. (Jha and Sinha, 2013)

Another problem in conducting a commodity price forecast is the volatility of

prices over time. A study regarding commodity price forecast resulted in forecast prices

increasing rapidly, and in the long-run becoming larger due to a spike in futures prices.

This resulted to a lower accuracy of the forecasts. It was also mentioned in the study

that in order to improve forecast accuracy, dummy variables may be used to adjust for

price spikes. Technically, it can be observed that there is a need to compare forecasting

models with the other models to ensure that a proper model is used in a proper scenario.

(Bowman and Husein, 2004)

Another study dealt with the problem of short-term market price forecasting.

Time series analysis is usually used in dealing with this problem. Furthermore, ANN, a
7

De La Salle University – Dasmariñas

new technique, has been discovered as a tool in price forecasting. In this study, ANN

model has been compared with the time series autoregressive integrated moving

average (ARIMA) in forecasting the price of tomato from years 1996 to 2010. The

results showed that ANN model performed better than ARIMA model in terms of their

relative errors. (Li, Xu and Li, 2010)

Corn is second to rice as the most important crop in the Philippines, and yet the

studies done regarding forecasting the price of corn in the Philippines are very few. We

can only see studies done about the pricing behavior of Philippine corn, relationship

between trade liberalization and Philippine corn prices, relationship between the prices

of Philippine rice and corn, socio-economic impact of corn in the Philippines, etc.

Basically, these studies only present behaviors, relationships, performances, impacts,

etc. Like in the other countries, it is important to emphasize methodologies for the

improvement of forecasting of the price of corn in the Philippines in order to aid both

the producers and consumers in making sound decisions. Specifically, this study

answered the following questions:

1. What is the trend of the monthly farmgate prices of corn in the Philippines from

2007 to 2017?;

2. How do Autoregressive Integrated Moving Average (ARIMA) and AR models

perform in forecasting the price of corn in the Philippines?; and

3. How do the predictive data and actual data differ?

Objectives of the Study


8

De La Salle University – Dasmariñas

Generally, this study aimed to provide a forecast on the price of corn in the

Philippines. In order to carry out the general objective in a more organized and

systematic way, the following specific objectives were made:

1. To describe the trend of the monthly farmgate prices of corn in the Philippines from

2007 to 2017;

2. To investigate the performances of ARIMA and AR models in forecasting the price

of corn in the Philippines; and

3. To analyze the differences between the predictive data and actual data.

Hypotheses of the Study

Dash, Solanki and S. (2012) conducted a study in India regarding commodity

market behavior, price and its factors. Included in these commodities are the three agro-

products, namely: channa, wheat and pepper. The main factor that affects the prices of

these crops, in terms of supply and production, is the monsoons. These crops are also

affected by storage constraints that are temporary. Other factors include inflation,

supply constraints, costs of production, foreign exchange holdings, and some

international policies pertaining to imports and exports. Thus, in order to carry out the

study more properly and systematically, the researcher hypothesized that:

H1: There is an existence of either trend or seasonality in the corn prices.

Teucrium Trading, LLC (2015), in one of its articles, stated that there is a

seasonal price patterns in corn. Some of the findings of the study include, but not

limited to, the following, namely: (1) the world’s two largest corn-producing countries,
9

De La Salle University – Dasmariñas

the US and China, supplies the biggest quantity of corn for approximately 12 weeks,

starting from mid-September to mid-December; (2) December is the month with the

greatest number of price decreases; and (3) the investors expect great opportunities

from both the US and China given the seasonal pattern of their corn prices.

H2: ARIMA model provides a better fit than AR model in forecasting the price

of corn.

Jadhav, Reddy and Gaddi (2017) conducted a study on the application of

ARIMA Model for forecasting the prices of paddy, ragi, and maize (corn) in India. The

results showed that ARIMA Model is a powerful tool in forecasting commodity prices.

Furthermore, the research checked the validity of the model using the values of MSE,

MAPE, and Theil’s U, and these values indicated that the forecasted values are almost

similar to the actual values. Lastly, one of the limitations of the ARIMA Model is that

the time series should be long, which makes the said model really suitable in

forecasting the price of corn in the Philippines.

Significance of the Study

This study compared the performances of both AR model and ARIMA model in

order to determine the model that is flexible enough to the volatility of corn’s prices in

the Philippines.

The government, most especially the policy makers, this impacts their decision

as to how they are going to forecast the price of corn in the Philippines. Given the
10

De La Salle University – Dasmariñas

uncontrollable circumstances that could negatively affect the commodity prices, it is

better to have many alternative models that could fit the scenario given certain factors.

The farmers are guaranteed to benefit on this study as they will be guided on

what decisions should be made in the future in order to be financially stable. Having a

reliable commodity price forecasting method to account for yields will be very helpful.

Though farmers are considered starving and dying in the Philippines, the opportunity to

receive financial incentives in the future is always there for as long as they are willing

to grab it.

The students should be able to learn the value of food security in the long-run as

early as possible. In response to this, through this study, they will learn that commodity

price forecasting is not simply about being able to understand numbers and figures, but

by those figures and numbers, policies can be derived in order to secure food in the

long-run.

This study could be further improved by the future researchers who will be

conducting a research similar to this. The fact that this study only has one variable, it

might be better for the other researchers to come up with models, aside from the

commonly used ARIMA and AR models, which could easily deal with univariate

analysis while also looking into the effectiveness of their performances as well.

Scope and Limitations


11

De La Salle University – Dasmariñas

This study covered the prices of corn from 80 provinces/cities including Metro

Manila, the same with the provinces/cities covered by the Philippine Statistics

Authority (PSA).

This study is limited only to the data available at PSA as the said organization

has the wholesale, retail, and farmgate prices of corn in the Philippines. This follows

the assumption that the data provided by PSA are all accurate.

This study is limited only to the use of two models, AR and ARIMA. This

paper’s main model will be ARIMA while AR will only be a model for comparison.

The data that used in forecasting the price of the corn in the Philippines is only

from 2007 to 2017 because the data from these years are still available and accessible

through the data sources of this study.

Definition of Terms

Commodity Price refers to the wholesale, retail, or farmgate price of crops such as rice,

corn, sugar, cassava, vegetables, fruits, and rubber, which could be either

wholesale or retail.

Corn or yellow corn specifically is the second most important crop in the Philippines,

and is the main subject of this study.

Farmgate Price means the price of corn set by the producer itself. It is also termed as

the producer price.

Forecasting is the method used in this study that uses historical prices of corn in order

to determine the gap between the actual and forecasted values.


12

De La Salle University – Dasmariñas

Price refers to the farmgate prices of corn in the Philippines.

CHAPTER II

REVIEW OF RELATED LITERATURE

This chapter discussed some past researches conducted that are related to

forecasting the price of corn in the Philippines. It started with the discussion of previous

literature regarding the impact of commodity prices to the economy, thus indicating the

economic impact of this study. This is then followed by the discussion of the factors
13

De La Salle University – Dasmariñas

that affect the price increases and decreases of commodities. Finally, presented in this

chapter are the methods of commodity price forecasting done by various researchers in

the past. The related studies done by the researchers in the past enabled the researcher

to assess and analyze the studies that have been conducted before, which created a

foundation for this study. Furthermore, the researcher determined what has been

discussed by the previous studies so far, and what has not yet been discussed that can

serve as a research gap.

Impact of Commodity Prices to the Economy

Sands (2015) stated that fluctuations in commodity prices affect the entire

economy in terms of employment, public and private expenditures, and capital

accumulation. When the prices fluctuate down, the rate of return of commodity sectors

exceeds that of the non-commodity sectors. In addition, a lot of economic problems

arise whenever economies rely on commodities as the main component of their Gross

Domestic Product (GDP). Because of this, we see a shift from commodity sectors into

productive non-commodity sectors. Brazil is said to be one of the major commodity

exporters all over the world, and it has its own major stocks as well. However, a

commodity deflation has been experienced at around April 2015, which forced Brazil’s

majors stocks to give negative returns. The researcher then concluded that in order to

adjust to lower commodity prices, two steps under fiscal policy can be undertaken. First

is for the government to reduce taxes to increase household spending. Last is to handle

both unemployment and the investment cycle by investing in other productive assets.
14

De La Salle University – Dasmariñas

The case of Pakistan clearly displays the relationship between commodity prices

and employment. Pakistan, which has a moderate amount of oil production, is a country

that relies heavily on oil imports to supply the oil demand most especially by the

industrial sector. Ahmad (2013), in his study about the effect of oil prices on

unemployment, stated that there has been a very few existing literature regarding the

relationship between unemployment and oil process in developing countries. This

proposed a challenge because Pakistan is a developing country which is damaged on

the increasing oil prices. The results indicated that there is a significant relationship

between oil prices and unemployment, but no significant relationship between real

interest rate and unemployment. Furthermore, it is suggested that there is a significant

relationship between real oil prices in Pakistan are significantly related to the real

interest rate. Finally, the study concluded that in the long-run, the oil prices can be used

to forecast unemployment rate and real interest rate.

An Australian economist said that it can be challenging on the part of a

researcher to analyze how commodity products are likely to impact both the customers

and the whole economy. The economist further explained that one of the pressing issues

concerning commodity markets is the dramatic declines in the industrial commodity

prices such as iron-ore and oil. Basically, the study showed how this scenario would

impact both the global and Australian economies. For the global economy, a fall in oil

prices will have significant implications for oil importers and exporters, consumers and

governments. In this case, Russia and Organization of the Petroleum Exporting

Countries (OPEC) countries, which rely heavily on oil revenues to fund their
15

De La Salle University – Dasmariñas

government expenditures, will lose a lot during heavy price decreases of oil. Although

affiliated companies such as energy-mining companies and the like will be negatively

affected, a lot of countries will still benefit. In fact, industries that have higher input

costs on oil will have free cash flows, and will be able to operate at higher margins. As

for the Australian economy, the results showed that the impacts will most likely be seen

in inflation and interest rates in the short-run. (Oster, 2015)

Aside from the industrial sector, an agricultural sector also plays a vital role in

determining a country’s economic development, most especially in developing

countries. Countries such as Liberia and Somalia account agriculture as more than 50%

of GDP. An agricultural sector becomes successful provided that it supports economic

growth. The US has a strong economy in terms of agriculture. American farmers are

capable of producing vegetables, fruits, grains, meat, and dairy products at a low cost.

As a result of this, domestic food supply becomes safe and secured. Furthermore,

through modern technology, the American agriculture sector is capable of producing

biofuels and other sources of alternative energy in order to minimize dependence on

foreign oil. This helps to reduce the costs incurred by the businesspeople and

consumers in purchasing gas or oil. Finally, it is truly important for rural areas and

small towns to have a strong agricultural economy. In fact, farmers and ranchers give

full support to farm industries, and they purchase local goods and services, which

results to an increased production. This high level of production has contributed a lot to

the businesses given that a strong agricultural economy exists. (United States Congress

Joint Economic Committee [JEC], 2013)


16

De La Salle University – Dasmariñas

Determinants of Commodity Prices

There has been a vast study regarding both short- and long-term determinants of

commodity prices. Over the years, studies pertaining to this topic become more

prevalent. Good (2008) conducted a study on the factors affecting corn and soybean

prices. The researcher stated that the agricultural commodities have been influenced by

the change of value of US-Dollar which has a negative relationship for both the corn

and soybean prices. Changes in crude oil prices are considered to affect both the corn

and soybean prices negatively. News pertaining to exports also affects both corn and

soybean prices. Weather is an important factor to every agricultural commodity, corn

and soybean included. Another important factor is production as it is highly related with

weather. Finally, the developments in the financial markets have positive effect on corn

and soybean prices. Similarly, any weakening of those markets will have a negative

effect on both commodities.

Determinants of commodity prices, which are either short- or long-run in nature,

can also be either microeconomic or macroeconomic. Frankel and Rose (2009),

explained that agricultural and mineral commodities peaked sharply in 2008. The main

causes included the ease in monetary policy due to the low real interest rates, a

speculative bubble which arose from expectations, some risks and uncertainties, and

strong global growth. This sharp spike which happened in 2008 led the researchers in

the study which resulted to the analysis of macroeconomic and microeconomic

determinants of 11 individual commodity prices. The results indicated that although the
17

De La Salle University – Dasmariñas

macroeconomic determinants: global GDP, and real interest rate both have a positive

relationship on real commodity prices, the microeconomic determinants: inventory

levels, uncertainty measures, and the spot-futures spread have the strongest effects on

real commodity prices. Additionally, there is an existence of bandwagon effect.

Similar to the study conducted by Good (2008), in a macroeconomic

perspective, the determinants of agricultural commodity price volatility include the

following: (1) stocks that has a negative relationship with price volatility ; (2) Southern

Oscillation Index (SOI) that has a positive relationship with price volatility; (3) world

market structure that has a negative relationship with price volatility; (4) biofuel

production that has a positive relationship with price volatility; (5) Kilian index; (6)

crude oil price behavior that has a positive relationship with price volatility; (7) US-

Dollar exchange rate volatility that has a positive relationship with price volatility; (8)

US interest rate that has a negative relationship with price volatility; (9) the Scalping

index; and (10) the Working-T index. The performances of Generalized Autoregressive

Conditional Heteroskedasticity-Mixed-data Sampling (GARCH-MIDAS) and

GARCH(1,1) were compared, which GARCH-MIDAS always performed better than

the other model. This analysis was applied and tested for wheat, corn, and soybean.

(Dönmez and Magrini, 2013)

Adeyanju (2014) argued that corn has been an important food to the entire

human race. However, more than just a food source, corn has also become an important

fuel source. Thus, the researcher enumerated the top factors that either increase or

decrease the price of corn. First is the effect of Ethanol, which comes from corn. Given
18

De La Salle University – Dasmariñas

that an increase in the demand for ethanol would increase the demand for corn, which

will surely increase the price of corn. However, when the demand for ethanol decreases,

decreasing the demand of corn, it is not necessarily equal to the effect of increasing

demand for corn given that only 40 percent of corn becomes ethanol. Another factor is

the crude oil prices which has a positive relationship with corn prices most of the time.

This is because even corn has been functional as an energy commodity as well. Next is

the speculator effect, which is considered to be the biggest driver of corn prices.

Naturally, it will be smart for investors to observe how corn is being valued before

taking any actions. Climate is a very important factor of corn included. Another

important factor, though not as significant as the other factors, is the Chinese effect.

China is said to be taking efforts to have a cleaner energy, therefore there will be an

increase in demand for ethanol, which will most likely contribute to an increase in

demand for corn. Finally, geopolitical issues play an important role in the corn since

corn production is unevenly distributed worldwide. Technically, a change in economy

affects corn industries.

Commodity Price Forecasting

There were a lot of researches done on the forecasting of commodity prices

using different econometric methods. Most researchers generally use either ARIMA

model, or VAR (Vector Autoregressive) model, for multivariate studies, or AR, for

univariate studies, in commodity price forecasting. In fact, Tripathi et al. (2014)

conducted a study in India regarding rice productivity and production using ARIMA
19

De La Salle University – Dasmariñas

models. The paper focused on the analysis of trend of rice area, production, and

productivity of Odisha as compared to India using data from years 1950 to 2009. It also

focused on forecasting the rice area, production, and productivity using ARIMA

models. It was found out that there is an increasing trend in productivity and production

for both India and Odisha, with Odisha having a lesser rate of increase than India. The

researchers believed that it is because of the low input in agricultural operations and

other biotic and abiotic factors. Overall, it was proved that ARIMA model can be

successfully used to forecast rice area, productivity, and production for both Odisha and

India in the coming years.

In a study pertaining to forecasting major fruit crops productions in Bangladesh,

Box-Jenkins ARIMA model was used. The study aimed to fit the Box-Jenkins ARIMA

model in forecasting three of the major fruit crops in Bangladesh namely: Mango,

Banana, and Guava. It was found out that for Mango, the best chosen Box-Jenkins

ARIMA model, accounting for more than 5% level of significance, is ARIMA(2,1,3);

for Banana, it is ARIMA(3,1,2); and for Guava, it is ARIMA(1,1,2). The researcher

concluded that given that these three models are capable of practically explaining the

situation, they are the best model to use in forecasting. The researcher further

recommended that these models can be used for decision-making by the researchers,

policymakers, businessmen, etc. Finally, this study concluded that Box-Jenkins ARIMA

model performs good in short-term forecasting. (Hamjah, 2014)

In addition to the usage of ARIMA model in forecasting commodity prices of

various places and periods, other researchers have forecasted commodity prices using
20

De La Salle University – Dasmariñas

regime-switching models, which this paper will also use in forecasting the price of corn

in the Philippines. Ubilava and Helmers (2011) conducted a study regarding the impact

of El Niño Southern Oscillation (ENSO) – a natural phenomenon characterized by wind

variations and changes in sea surface temperature – on predicting world Cocoa prices.

The researchers contributed to the previous knowledge of commodity price forecasting

by considering that a nonlinear causal relationship between ENSO and world Cocoa

prices would be possible to compare the performances between linear and nonlinear

models. The smooth transition autoregressive framework (STAR) model, the model

used by the researchers, and is under the regime-switching models, proved that

nonlinear models are more reliable in out-of-sample forecasting compared to linear

models. Furthermore, the study concluded that there exists a Granger causality between

ENSO and world Cocoa prices.

The STAR model was also used in forecasting Corn and Soybean basis using

regime-switching models, a study conducted by Sanders and Baker (2012). In this

study, it was stated that producers of corn and soybean in the core production areas in

the US have noticed a great increase in the volatility of prices in their recent years,

which resulted to an increase of price risk of producers in decision-making. This paper

aimed to apply regime-switching models to formulate a model that could adjust to the

prices’ changing volatilities, and to provide more accurate forecasts especially in

periods of changing volatilities. The researchers found out over the course of their study

that time series econometrics perform better at short-term forecasting, but difficult to

use in long-term forecasting. Finally, the study concluded that regime-switching models
21

De La Salle University – Dasmariñas

do not provide real forecasting improvement over ARIMA models despite of statistical

significance in favour of the regime-switching models.

Aside from ARIMA and VAR models, and any other related models, there is yet

another method in forecasting commodity prices that is not commonly used, but

according to the literature, this method has been used a couple of times in various

fields. The artificial neural network (ANN) is one of the useful tools in machine

learning that was once limited to studies pertaining to brain and psychology, but is now

used in variety of topics such as business, education, arts, and many more. Kulkarni and

Haidar (2009) developed a forecasting model for crude oil price on the basis of ANN

and commodity futures prices. Similar to the ARIMA model, there is also an optimal

ANN model structure where a researcher needs to be very careful about. The study

dealt with pre-processing the spot and futures prices into a couple of months in order to

determine the optimal lag. The process resulted to a model with an optimal lag of 13

lags to conduct a short-term forecast of until three days in the future. The forecast

accuracy of this model for the first, second, and third days are 78%, 66%, and 53%,

conclusively. This model is expected to help in the further understanding of crude oil

prices, and to enable the investors to have an effective risk management.

This paper focused on forecasting the price of corn in the Philippines. The

previous studies that have been presented in this section clearly explained the need to

forecast commodity prices in various places, as well as how these commodity prices

will have an impact on the economy. Through these past studies done by different

researchers, this paper was able to contribute additional knowledge in commodity price
22

De La Salle University – Dasmariñas

forecasting by formulating a methodology that will forecast the price of corn in the

Philippines.

CHAPTER III

FRAMEWORKS OF THE STUDY

Theoretical Framework

The previous chapters of this study have mentioned some among the numerous

studies done on commodity price forecasting models. As mentioned in the Chapter II of

this study, among the most used models by the researchers when dealing with

commodity price forecasting are ARIMA and AR/VAR models. Though not as common

as the previous mentioned two models, there are still a lot of models that can be used in

forecasting commodity prices as they will have their own importance depending on the

scenario.

Judgmental forecasting. This a forecasting method which relies on a person’s

own judgment of a particular situation. It is naturally expected to be subjective because

it does not rely on historical and other statistical data, which means that it can only be

used on qualitative researches. Hillier, F. S. and Hillier, M. S. (2001) enumerated the

commonly used judgmental forecasting methods, namely: (1) manager’s opinion which
23

De La Salle University – Dasmariñas

relies on a single manager’s best judgment in forecasting; (2) jury of executive opinion

that is similar to the first one, except now that there is a small group of managers who

combine their best judgments; (3) sales force composite which is often used by the

companies when they want to generate higher sales by hiring sales forces; (4) consumer

market survey that relies on surveying actual or potential customers in order to

determine their responsiveness to the new products or new features of the existing

products; and (5) Delphi method which involves a group of experts from various

locations independently filling out a series of questionnaires.

Unit root model. The unit root problem is demonstrated when the presence of

unit root in a time series affects statistical inferences due to some vague, unpredictable

patterns. The solution provided to this problem is the unit root testing which ensures

that the time series is stationary, that is the statistical properties do not change over

time. Some commonly used unit root tests include, but not limited to, the Dickey Fuller

Test, Augmented Dickey-Fuller (ADF) Test, and Phillips-Perron (PP) Test. The unit

root model with trend and drift is the simplest form of forecasting model, and it can be

written as:

yt = µ + yt-1 + ut,

where yt is the natural logarithm of the commodity price at period t, and the error term,

ut is assumed to be a white noise.

ARIMA model. The ARIMA model was first introduced by the statisticians

George E.P. Box and Gwilym M. Jenkins and thus being commonly known as Box-

Jenkins model which is used as a forecasting model. This is probably the most
24

De La Salle University – Dasmariñas

commonly used model in forecasting commodity prices specified by the three order

parameters (p, d, q), and is also the most commonly used model in forecasting other

prices given that it can convert non-stationary time series data in to stationary time

series data using differentiation.

The equation for ARIMA model in a stationary time series analysis is a linear

equation, which can be expressed as:

Futures forecast model. The futures price is one way of forecasting commodity

spot prices. Mckenzie and Holt (1998) and Chinn and Coibion (2010) stated that the

futures price is an unbiased predictor of future spot prices, and there is a little evidence

that it is also the best forecast according to Alquist and Kilian (2010) and Alquist et al.

(2011). Despite of a large literature proving that the capacity of futures price to forecast

exceeds that of the random walk model, the model concerning futures prices performs

differently depending on the commodity, whether it is consumed daily, weekly,

monthly, or even yearly. The general futures forecast model is expressed as:

St = α + βFt|t-k + et,
25

De La Salle University – Dasmariñas

where Ft|t-k is the price for period t with future markets in period t-k.

Vector autoregressive model. The VAR Model, which is a simple, yet flexible

model that deals with multivariate time series data, is just a natural extension of the AR

Model, which deals with univariate time series data. Being one of the most commonly

used model in forecasting commodity prices, VAR Model is often compared to ARIMA

Model alongside Error Correction Model (ECM) in terms of their effectiveness given

various situations. However, it was also found out that there are times when VAR

Model is preferred over ARIMA Model because there are more theoretical backgrounds

on the former model than the latter. This model was popularized by the American

econometrician and macroeconomist Christopher A. Sims (1980) on his journal article

entitled Macroeconomics and Reality. In that article, Sims demonstrated that VAR

model is able to provide a flexible, better framework in analyzing economic time series

data. Assuming there are three different time series variables, denoted by xt,1, xt,2, and

xt,3, the VAR model of order 1 is expressed as:

xt,1 = α1 + ϕ11xt-1,1 + ϕ12xt-1,2 + ϕ13xt-1,3 + wt,1

xt,2 = α2 + ϕ21xt-1,1 + ϕ22xt-1,2 + ϕ23xt-1,3 + wt,2

xt,3 = α3 + ϕ31xt-1,1 + ϕ32xt-1,2 + ϕ33xt-1,3 + wt,3 ,

where α is constant, ϕ is the phi coefficient, and wt is the error term.

Conceptual Framework

Mentioned in the hypotheses of the study are the characteristic and trend of the

commodity prices, most especially price of corn. Furthermore, it has been mentioned
26

De La Salle University – Dasmariñas

the superiority of farmgate prices over wholesale and retail prices, and the importance

of focusing more on ARIMA model than the other models. Figure 1 represents

specifically the model which this study used in forecasting the price of corn in the

Philippines.
Autore
Autore
gressiv
gressiv
ee
Model
Model

Pre
dict
ed
Actual Far
Farmgat mga
e Corn te
Prices Cor
n
Pric
es
Autore
Autore
gressiv
gressiv
ee
Integrat
Integrat
ed
ed
Moving
Moving
Averag
Averag
ee
Model
Model
De La Salle University – Dasmariñas

Figure 1. Overall framework of the research


27
28

De La Salle University – Dasmariñas

CHAPTER IV

METHODOLOGY

Research Design

This study dealt with the quantitative aspect of research. Specifically, this paper

aimed to assess whether what model performs the best in forecasting the price of corn

in the Philippines. The models included ARIMA model and AR model. This study used

the historical design of research. The historical design of research enabled the

researches to gather and synthesize past data in order to accept or reject a hypothesis –

to prove whether corn prices in the Philippines have an upward or downward trend.

Furthermore, this study is also an evaluative research. This paper also provided

an evaluation and assessment on what model performs the best in forecasting the price

of corn in the Philippines. Since food security is a very serious matter not only in the

Philippines, but in the other countries as well, the forecasting method should be ensured

that it provides the best, most accurate forecasts as possible.

Sources of Data

This study gathered data from the secondary sources that are available and

accessible to the public online. These data came from government agencies, specifically

the Philippine Statistics Authority (PSA), whose scope includes the gathering price of

the agricultural crops in the Philippines.


29

De La Salle University – Dasmariñas

Methods of Data Analysis

This study used some statistical techniques depending on the requirements

presented on the objectives of the study. This section mentioned the different statistical

techniques that this study employed. In the case of historical design, tables and graphs

are used in order to clearly see the trend of prices of corn in the Philippines. Using

these tools enabled the researcher to analyze the patterns displayed in the historical data

gathered, which will led to an intelligent conclusion as to why such pattern/s occurred.

As to the evaluative design of this study, both the ARIMA and AR models are

chosen for comparison as to what model performs best in forecasting the price of corn

in the Philippines. These two models are suitable to use when a particular study

concerning forecasting has only one variable available.

In the case of ARIMA forecasting model, six steps will be followed for a more

comprehensive model, namely: (1) examining the data, where patterns and

irregularities are checked, outliers and missing values are properly filled in, and

converting the prices into logarithmic form to better fir the model; (2) decomposing the

data, wherein the seasonal, trend, and cycle components are removed; (3) stationarity

testing, where the prices are checked whether they have unit root problem or not; (4)

test for autocorrelation and the selection of the best ARIMA model by checking the

autocorrelation function (ACF) and partial ACF (PACF); (5) fitting the selected ARIMA

model; and (6) evaluation of the performance of ARIMA model. (Dalinina, 2017)
30

De La Salle University – Dasmariñas

In order to determine the significance of the overall models of the study, both

the coefficient of determination (R2) and the F-statistic are checked as well. Eviews is

used in the estimation procedure.

The ARIMA model, which satisfied the second objective of the study is:

Ŷ = µ + Yt-1

or

Predicted Value of FPRICE = µ + FPRICEt-1 ,

where:

FPRICE = Farmgate Price of Corn in the Philippines (in PhP/kg.)

t = Time

µ = Constant term, average change over time

The AR model, on the other hand, which satisfied the third objective of the

study is:

Ŷ = ϕYt−1 + ut

or

Predicted Value of FPRICE = ϕFPRICEt−1 + ut ,

where:

ϕ = Phi coefficient (should not be less than 1)

u = Random error at period t

There are various ways of measuring the effectiveness of forecast performances

of different models. Hyndman (2014), in his study on measuring forecast accuracy,

explained three points on why researchers should rely on measures on forecast accuracy
31

De La Salle University – Dasmariñas

in forecasting prices, namely: (1) a model that perfectly fits the actual data does not

necessarily perform well in forecasting; (2) with sufficient number of parameters,

perfect fit can be obtained; and (3) over-fitting the model to a data is not a good idea

because it fails to understand the systematic pattern of the data. Furthermore, he

enumerated various measures of forecast accuracy which the researches can rely on

when forecasting prices. Under scale-dependent errors, the two most commonly used

measures are Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). In

percentage errors, there is only Mean Absolute Percentage Error (MAPE). Lastly, under

scaled errors, there is only Mean Absolute Squared Error (MASE).

There was a study conducted by Hyndman and Koehler (2005) regarding a

closer look at the measures of forecast accuracy. It was mentioned in the study that the

Theil’s Inequality Coefficient or Theil’s U Statistic is also a good measure of forecast

accuracy that is commonly cited in literature reviews.

This research primarily focused on using RMSE, Theil’s Inequality Coefficient,

and MAPE in checking the forecast errors of both AR and ARIMA. RMSE is also a

suitable measure to use given that it can only be used for a specific commodity and not

for comparison across various commodities. MAPE was chosen over MAE because in

some cases, the exact values don’t clearly tell whether the error has significant

difference or not. Thus, percentages are far more trusted than actual values.

The formula for RMSE is written as:

RMSE = √ mean( e 2i ) ,
32

De La Salle University – Dasmariñas

where ei is the forecast error expressed as the difference between the actual and

predicted values in the forecast sample at period i.

The general formula of Theil’s Inequality Coefficient is expressed as:


n
1
n
∑ e2
i =1
TH = ,

√ √∑
n n
1
n
∑ y + 1n2
ŷ 2

i =1 i =1

where n is the sample size of the study, ŷ is the predicted value of y, and e is the

equivalence factor, denoting economies of scale.

The general formula for MAPE is written as:


n
1
MAPE =
n
∑ || Actual
Actual |
- Forecast |
x 100
i =1

CHAPTER V
33

De La Salle University – Dasmariñas

RESULTS AND DISCUSSION

The first part of this section provided the monthly farmgate prices of corn in the

Philippines from years 2007 to 2017, which were obtained primarily from PSA. It is

followed by the results of ARIMA and AR models pertaining to their respective forecast

performances in the farmgate prices of corn in the Philippines, as well as the detailed

discussions of those results. Furthermore, this chapter provided a decision criteria on

which model performed better in terms of forecasting the farmgate prices of corn in the

Philippines. The last part of this chapter compared the predictive and actual data using

both the AR and ARIMA models.

Timmer (2008) conducted a study concerning the causes of high food prices. It

is because of these high food prices that poor consumers are experiencing grave

consequences concerning food security. The study concluded some factors that affect

the food prices depending on the year. In 2004, at least three main factors are found to

be dominant, namely: (1) China’s rapid economic growth and the excess of demand

over supply in India; (2) a constant decline in the value of US dollar; and (3) the

combined high and still rising prices of fuel that were found out to be related to the

other commodity prices.

In the Philippines, one of the most common agricultural problems is the climate

or weather. During typhoons, the usual scenario is that people expect a price spike in

the agricultural prices due to the damage dealt to the farmlands and its farmers.

Contrary to this belief, the Bureau of Agricultural Statistics (BAS) (2013) stated that
34

De La Salle University – Dasmariñas

prices of rice and corn remained stable in Visayas region during the week when

typhoon Yolanda, one of the strongest typhoons recorded in the world, devastated the

said region.

This is a scenario which is not commonly seen among different countries, and

therefore should not be expected to frequently occur. Padin (2016) reported that the

average farmgate prices of local corn have risen during the recent weeks as El Niño

continues to pester the areas in the Philippines where corn is thriving. Here, the farmers

had a difficult time earning due to the harsh climate, which forced the prices of corn to

increase.

Trend of Farmgate Prices of Corn in the Philippines

Tiffany (2009) conducted a study in the US pertaining to the environmental and

economic impacts of the usage and production of US corn ethanol. In this paper, the

objectives include, but not limited to, the impacts of corn ethanol production to the

farmers’ decision-making, and the relationship of corn ethanol production and the

prices of corn. Wisner (2014) showed the positive relationship of crude oil and corn oil

prices from 2003 to 2014. It led to a conclusion that corn ethanol, being a substitute of

crude oil, also led to an increase in the corn prices. In the Philippines, various factors

are considered when corn prices. Table 1 shows the monthly farmgate prices of corn in

the Philippines from 2007 to 2017 while Figure 2 is the presentation of these tabulated

prices in a graphical form. Generally, from the graph, the trend is found to be upward,

coupled with evident price fluctuations. The upward trend, with some price
35

De La Salle University – Dasmariñas

fluctuations, can be attributed to a lot of factors. Padin (2016) proved that natural

weather, no matter how unlikely the result is, can affect the corn prices in the

Philippines. Juliano and Gonzales (n.d.) mentioned technological advancement and

policies as important factors in the movement of corn prices. With technological

advancements, it becomes much easier for the farmers to produce more corn, and for

them to decide the farmgate price of corn. Furthermore, the study also mentioned that

policies regarding low taxes on fertilizers, and credit at reasonable rates will definitely

enhance corn production.

Table 1
36

De La Salle University – Dasmariñas

Monthly Farmgate Prices of Corn, Philippines, 2007-2017

Year Month Price Percent change


(in PhP/kg.)
2007 January 9.28 -
February 9.62 3.66%
March 10.02 4.16%
April 10.31 2.89%
May 10.65 3.30%
June 10.51 -1.31%
July 10.26 -2.38%
August 10.07 -1.85%
September 9.81 -2.58%
October 9.93 1.22%
November 9.90 -0.30%
December 9.98 0.81%
2008 January 10.24 2.61%
February 10.60 3.52%
March 10.84 2.26%
April 11.04 1.85%
May 10.74 -2.72%
June 11.04 2.79%
July 10.79 -2.26%
August 10.51 -2.59%
September 10.45 -0.57%
October 10.84 3.73%
November 11.28 4.06%
December 11.71 3.81%
2009 January 12.91 10.25%
February 13.55 4.96%
March 12.38 -8.63%
April 12.07 -2.50%
May 11.51 -4.64%
Continued

37

De La Salle University – Dasmariñas

Year Month Price Percent change


(in PhP/kg.)
June 10.35 -10.08%
July 10.05 -2.90%
August 8.99 -10.55%
September 8.49 -5.56%
October 8.68 2.24%
November 9.49 9.33%
December 9.71 2.32%
2010 January 10.13 4.33%
February 10.82 6.81%
March 11.40 5.36%
April 11.30 -0.88%
May 11.02 -2.48%
June 11.40 3.45%
July 11.31 -0.79%
August 10.96 -3.09%
September 10.38 -5.29%
October 10.55 1.64%
November 11.19 6.07%
December 11.92 6.52%
2011 January 12.55 5.29%
February 12.85 2.39%
March 12.59 -2.02%
April 11.74 -6.75%
May 11.66 -0.68%
June 11.55 -0.94%
July 11.32 -1.99%
August 10.89 -3.80%
September 10.73 -1.47%
October 11.65 8.57%
November 12.69 8.93%
Continued

38

De La Salle University – Dasmariñas

Year Month Price Percent change


(in PhP/kg.)
December 13.01 2.52%
2012 January 13.67 5.07%
February 13.58 -0.66%
March 12.63 -7.00%
April 12.46 -1.35%
May 12.72 2.09%
June 12.46 -2.04%
July 12.32 -1.12%
August 12.14 -1.46%
September 12.01 -1.07%
October 12.28 2.25%
November 12.56 2.28%
December 12.58 0.16%
2013 January 12.48 -0.79%
February 11.97 -4.09%
March 11.87 -0.84%
April 11.89 0.17%
May 12.02 1.09%
June 12.08 0.50%
July 11.83 -2.07%
August 11.58 -2.11%
September 11.74 1.38%
October 11.81 0.60%
November 11.75 -0.51%
December 11.83 0.68%
2014 January 11.97 1.18%
February 11.94 -0.25%
March 12.21 2.26%
April 12.54 2.70%

Continued

39

De La Salle University – Dasmariñas

Year Month Price Percent change


(in PhP/kg.)
May 13.02 3.83%
June 13.42 3.07%
July 13.22 -1.49%
August 13.94 5.45%
September 12.95 -7.10%
October 12.84 -0.85%
November 12.51 -2.57%
December 12.50 -0.08%
2015 January 12.28 -1.76%
February 12.40 0.98%
March 12.66 2.10%
April 12.67 0.08%
May 12.79 0.95%
June 12.68 -0.86%
July 12.65 -0.24%
August 12.40 -1.98%
September 11.82 -4.68%
October 11.60 -1.86%
November 11.49 -0.95%
December 11.43 -0.52%
2016 January 11.67 2.10%
February 12.25 4.97%
March 12.60 2.86%
April 12.51 -0.71%
May 12.87 2.88%
June 13.00 1.01%
July 12.98 -0.15%
August 12.48 -3.85%
September 11.54 -7.53%
October 10.96 -5.03%
Continued

40

De La Salle University – Dasmariñas

Year Month Price Percent change


(in PhP/kg.)
November 10.86 -0.91%
December 10.96 0.92%
2017 January 11.25 2.65%
February 11.45 1.78%
March 11.41 -0.35%
April 11.41 0.00%
May 11.53 1.05%
June 11.33 -1.73%
July 11.20 -1.15%
August 11.21 0.09%
September 11.43 1.96%
October 11.58 1.31%
November 12.07 4.23%
December 12.34 2.24%
Mean 11.60 0.28%
Standard Deviation 1.08 3.65%
Source: Philippine Statistics Authority
41

De La Salle University – Dasmariñas

Figure 2 shows the graphical representation of the monthly farmgate prices of

corn in the Philippines. The highest price, which is 13.94 PhP/kg., occurred on August

2014. The PSA has attributed this to the high corn output on that same year. This good

corn production, which led to high farmgate prices, was caused by the increased use of

good quality seeds. Despite of the high farmgate prices in 2014, there was a downward

trend of prices in 2015 due to the lingering effects of El Niño, as well as the shortage of

water supply due to drought. The lowest price, 8.49 PhP/kg., occurred on September

2009. This is primarily because 2009 was considered the deadliest season in the

Philippines after decades. According to Corpuz (2010), typhoons Ketsana (Ondoy) and

Parma (Pepeng), which both occurred on this year, have devastated a lot of Filipino

homes as well as the entire Philippine agriculture, with its effects remaining until 2011.

Consequently, some price drops in 2010 can be attributed to the occurrence of El Niño.

Though there are a few rare instances where natural calamities don’t necessarily

negatively affect agriculture, this is usually an expected outcome especially to the

farmers.
42

De La Salle University – Dasmariñas

2017
2016
2015
2014
Farmgate Prices of Corn

2013

Ye ar
2012
2011
Figure 2. Farmgate prices of corn in the Philippines

2010
2009
2008
2007
11.00

9.00

8.00
14.00

13.00

12.00

10.00

Farmgate Price (in PhP/kg.)

Theories have stated that commodity prices are bound to increase over time due

to inflation. Regarding the price fluctuations, it is not just the frequency of them that
43

De La Salle University – Dasmariñas

matters, it should also be important to consider how high or low the price increases and

decreases are. In a study conducted by Bäckman and Sumelius (2009), there are

numerous factors affecting the price fluctuations of food products according to various

literature reviews. The researchers enumerated some common and a few uncommon

factors considering both the supply and demand factors. Under demand factors, there

are three, namely: (1) energy price, which has a positive relationship with the demand

of agricultural products; (2) population growth, which exhibits a positive relationship

on the demand for agricultural commodities, and is the least emphasized one among the

demand factors; and (3) consumer habits, which positively affects the demand for meat

products, and is defined by the study as the increased protein intake of the consumers.

There are six enumerated supply factors in the study, namely: (1) input factors, which

directly affect the production of agricultural products; (2) weather, which refers to the

natural occurrence that humans cannot control such as rain, or even floods, typhoons,

and the presence of insects in the farm; (3) climate, which is capable of changing the

agricultural production from one place to another; (4) technological development,

which exhibits a positive relationship on the supply of goods, that drives increased

production; (5) policies and institutions, which means the unstable policies, rules, and

regulations regarding the commodity production, that negatively affects the supply of

agricultural foods; and (6) prices, which have a direct relationship on the supply of

commodities.

Performance of the Forecasting Models


44

De La Salle University – Dasmariñas

This is the second part of this section where the researcher conducts an

evaluation as to what model is better in forecasting the price of corn in the Philippines.

As mentioned in Chapter IV, there will be six steps to be followed under the ARIMA

model. Furthermore, mentioned in the Methodology section, are the bases of this study

in determining the best model, which are the values of MAPE, RMSE, and Theil’s

Inequality Coefficient.

Autoregressive Integrated Moving Average Model

Step 1: Examining the Data

Table 1 and Figure 2 showed the monthly farmgate prices of corn in the

Philippines in tabular and graphical forms, respectively. It can be seen that there are

complete data from 2007 to 2017, without any missing values. Due to the volatility of

prices of corn, outliers are expected to occur, like what is shown above. It is worth

noting some comparisons between the logarithmic form of farmgate prices compared to

its original form. First, the logarithmic prices show that they have been deflated as

compared to the original prices. Second, from the original trend of volatile price prices

with heavy price spikes and drops, it has now become a steady, an almost horizontal

trend with some weaker price spikes and drops. Lastly, the logarithmic form of

farmgate prices has helped in stabilizing the price increases and decreases.

Step 2: Decomposing the Data

Since the prices of corn are gathered monthly, it is now possible to decompose

the logarithmic corn prices in order to eliminate the presence of seasonality, trend, or

cycle. The researcher used the additive model, which is done by subtracting the
45

De La Salle University – Dasmariñas

seasonal component from the original series, in decomposing the entire series. This

resulted in a series having lower prices compared to the original series. Having a de-

seasonalized series means that the seasonal, cycle, and trend components of the series

have been eliminated. The next step required is to determine whether there is a

stationarity problem or not.

Step 3: Stationarity Testing

It is commonly assumed that a time series data is stationary. Stationarity refers

to the statistical properties, such as mean, variance and autocorrelation, which do not

change over time. Table 2 shows the Augmented Dickey-Fuller unit root test on de-

seasonalized logarithmic form of farmgate prices of corn (LFPRICESA) at level. The

results showed a Philips-Perron test statistics probability value of 0.9%, which is less

than 5%. This only means that the null hypothesis can be rejected, and that there is no

stationarity problem in the series.

Table 2

Augmented Dickey-Fuller Testing at Level

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.541718 0.0083


Test critical values: 1% level -3.481217
46

De La Salle University – Dasmariñas

5% level -2.883753
10% level -2.578694

*MacKinnon (1996) one-sided p-values.

Null Hypothesis: LFPRICESA has a unit root


Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=12)

Step 4: Autocorrelation Testing and ARIMA Model Selection

Table 3 shows the information on correlogram of LFPRICESA, with 36 lags

included. Given below are the values of autocorrelation (AC), partial autocorrelation

(PAC), Q-statistics and its p-values. AC does three important things, namely: (1) it

shows the correlation of the LFPRICESA values and lags; (2) it determines the

appropriate differencing of the best ARIMA model; and (3) it helps to determine the

order of the MA(q) component of the ARIMA model. Furthermore, the two lines in the

AC plot determine whether the autocorrelation is statistically different from 0 at 5%

significance level. If the AC is within the range of those two lines, then it is not
47

De La Salle University – Dasmariñas

statistically different from 0. On that note, all AC values at orders 1 to 36 are display

high autocorrelation that decays over time. On the other hand, PAC shows the

correlation of LFPRICESA values and its lags that is not explained by the past lags. It

also determines the order of the AR(p) component of the ARIMA model. Additionally,

there are also two lines in the PAC plot that function the same as the two lines in the

AC plot. It can be seen that the PAC value at order 1 displays a high autocorrelation

that cuts down to the next order. The Q-statistics values help determine whether the

values of LFPRICESA have correlation with each other or not. However, it is not easy

to manually choose the best ARIMA model to use given the combination of orders to be

tested.

Table 3

Correlogram of LFPRICESA
48

De La Salle University – Dasmariñas

Date: 05/10/18 Time: 11:45


Sample: 2007M01 2017M12
Included observations: 132

Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.916 0.916 113.26 0.000


2 0.790 -0.301 198.25 0.000
3 0.658 -0.055 257.62 0.000
4 0.530 -0.046 296.44 0.000
5 0.426 0.068 321.76 0.000
6 0.355 0.076 339.45 0.000
7 0.317 0.085 353.67 0.000
8 0.309 0.088 367.31 0.000
9 0.326 0.091 382.64 0.000
10 0.348 0.002 400.24 0.000
11 0.348 -0.114 417.92 0.000
12 0.315 -0.111 432.56 0.000
13 0.269 0.029 443.35 0.000
14 0.224 0.055 450.86 0.000
15 0.194 0.108 456.57 0.000
16 0.189 0.095 462.02 0.000
17 0.195 -0.019 467.86 0.000
18 0.221 0.082 475.46 0.000
19 0.257 0.000 485.83 0.000
20 0.288 -0.028 498.91 0.000
21 0.309 0.025 514.10 0.000
22 0.309 -0.003 529.50 0.000
23 0.282 -0.051 542.40 0.000
24 0.230 -0.074 551.06 0.000
25 0.187 0.102 556.85 0.000
26 0.164 0.059 561.32 0.000
27 0.146 -0.065 564.91 0.000
28 0.153 0.103 568.88 0.000
29 0.177 0.013 574.28 0.000
30 0.202 -0.024 581.36 0.000
31 0.224 -0.002 590.11 0.000
32 0.224 -0.100 598.96 0.000
33 0.208 0.046 606.70 0.000
34 0.164 -0.102 611.58 0.000
35 0.115 0.048 614.01 0.000
36 0.064 -0.081 614.76 0.000

Step 5: Fitting the ARIMA Model


49

De La Salle University – Dasmariñas

Table 4 shows the ARIMA model summary and criteria table for LFPRICESA.

In the criteria table, there are values of log-likelihood (LogL), Akaike Information

Criterion (AIC), Schwarz Criterion (BIC), and the Hannan-Quinn Criterion (HQ).

These three criteria are only used to determine the appropriate number of AR(p) and

MA(q) terms. The criteria table below chose ARIMA Model (3,3)(0,0) as the best

model given the lowest value of AIC, as well as the remaining criteria. This suggests

that the ARIMA Model, with the terms AR(3) and MA(3) without differencing, is the

best model to forecast the price of corn in the Philippines. However, this model will

still undergo further testing to see whether it satisfies the certain conditions to be met in

forecasting.

Table 4

ARIMA Criteria Table and Summary


50

De La Salle University – Dasmariñas

Model LogL AIC* BIC HQ

(3,3)(0,0) 292.026579 -4.303433 -4.128718 -4.232437


(4,2)(0,0) 291.828095 -4.300426 -4.125710 -4.229429
(3,4)(0,0) 292.459581 -4.294842 -4.098287 -4.214971
(4,3)(0,0) 292.287075 -4.292228 -4.095674 -4.212358
(3,2)(0,0) 289.608300 -4.281944 -4.129068 -4.219822
(4,4)(0,0) 292.463764 -4.279754 -4.061360 -4.191009
(2,4)(0,0) 287.438595 -4.233918 -4.059203 -4.162922
(2,0)(0,0) 283.022405 -4.227612 -4.140255 -4.192114
(2,1)(0,0) 283.297938 -4.216635 -4.107438 -4.172263
(3,0)(0,0) 283.286081 -4.216456 -4.107259 -4.172083
(4,1)(0,0) 285.083540 -4.213387 -4.060511 -4.151265
(3,1)(0,0) 283.999839 -4.212119 -4.081082 -4.158872
(4,0)(0,0) 283.352419 -4.202309 -4.071273 -4.149062
(2,2)(0,0) 283.305837 -4.201604 -4.070567 -4.148356
(1,4)(0,0) 284.282916 -4.201256 -4.048380 -4.139135
(2,3)(0,0) 284.093237 -4.198382 -4.045507 -4.136261
(1,3)(0,0) 283.057686 -4.197844 -4.066807 -4.144597
(1,2)(0,0) 281.812034 -4.194122 -4.084925 -4.149749
(1,1)(0,0) 280.017613 -4.182085 -4.094727 -4.146587
(1,0)(0,0) 269.112200 -4.032003 -3.966485 -4.005379
(0,4)(0,0) 267.913519 -3.968387 -3.837350 -3.915139
(0,3)(0,0) 254.651094 -3.782592 -3.673395 -3.738220
(0,2)(0,0) 235.022074 -3.500334 -3.412977 -3.464836
(0,1)(0,0) 196.542821 -2.932467 -2.866949 -2.905843
(0,0)(0,0) 126.106301 -1.880398 -1.836720 -1.862649

Model Selection Criteria Table


Dependent Variable: LFPRICESA
Date: 05/10/18 Time: 12:05
Sample: 2007M01 2017M12
Included observations: 132
51

De La Salle University – Dasmariñas

Table 5 shows the correlogram table of LFPRICE, after differencing once. As

opposed to the results of Table 3, this table shows that there is no significant

autocorrelation on both the AC and PAC plots, as shown by the values being within the

limit of the two lines in their respective plots. This means that there is no presence of

cuts or decays, and that the values under AC and PAC plots are all statistically the same

with 0 at 5% significance level.


52

De La Salle University – Dasmariñas

Table 5

Correlogram of D(LFPRICESA)
53

De La Salle University – Dasmariñas

Date: 05/10/18 Time: 11:51


Sample: 2007M01 2017M12
Included observations: 131

Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.397 0.397 21.124 0.000


2 0.153 -0.005 24.304 0.000
3 0.029 -0.036 24.420 0.000
4 -0.125 -0.147 26.551 0.000
5 -0.282 -0.217 37.556 0.000
6 -0.322 -0.157 52.008 0.000
7 -0.261 -0.080 61.571 0.000
8 -0.195 -0.077 66.954 0.000
9 -0.037 0.039 67.144 0.000
10 0.151 0.114 70.424 0.000
11 0.265 0.111 80.630 0.000
12 0.106 -0.173 82.278 0.000
13 0.022 -0.124 82.348 0.000
14 -0.112 -0.201 84.213 0.000
15 -0.203 -0.130 90.420 0.000
16 -0.179 0.019 95.248 0.000
17 -0.127 0.049 97.726 0.000
18 -0.108 -0.029 99.520 0.000
19 0.061 0.101 100.11 0.000
20 0.156 -0.028 103.94 0.000
21 0.225 -0.016 111.96 0.000
22 0.299 0.087 126.28 0.000
23 0.276 0.131 138.53 0.000
24 0.039 -0.092 138.77 0.000
25 -0.129 -0.056 141.50 0.000
26 -0.225 -0.108 149.91 0.000
27 -0.273 -0.100 162.36 0.000
28 -0.251 -0.024 173.00 0.000
29 -0.187 -0.008 178.95 0.000
30 -0.007 0.074 178.96 0.000
31 0.027 -0.046 179.09 0.000
32 0.101 -0.089 180.87 0.000
33 0.269 0.065 193.74 0.000
34 0.178 -0.078 199.45 0.000
35 0.025 -0.036 199.57 0.000
36 -0.101 -0.085 201.45 0.000

Table 6

ARIMA Criteria Table and Summary


54

De La Salle University – Dasmariñas

Model LogL AIC* BIC HQ

(2,3)(0,0) 289.712029 -4.283516 -4.130640 -4.221394


(3,4)(0,0) 291.260057 -4.276668 -4.080113 -4.196797
(2,4)(0,0) 289.787562 -4.269509 -4.094793 -4.198512
(3,3)(0,0) 289.777055 -4.269349 -4.094634 -4.198353
(4,2)(0,0) 289.775437 -4.269325 -4.094610 -4.198329
(4,4)(0,0) 290.940821 -4.256679 -4.038285 -4.167934
(4,3)(0,0) 289.796609 -4.254494 -4.057939 -4.174623
(2,2)(0,0) 284.959874 -4.226665 -4.095628 -4.173418
(2,1)(0,0) 283.656726 -4.222072 -4.112875 -4.177699
(4,1)(0,0) 285.496100 -4.219638 -4.066762 -4.157516
(3,1)(0,0) 284.467090 -4.219198 -4.088162 -4.165951
(3,2)(0,0) 284.961370 -4.211536 -4.058660 -4.149414
(1,4)(0,0) 283.396888 -4.187832 -4.034956 -4.125710
(1,3)(0,0) 282.002414 -4.181855 -4.050818 -4.128608
(1,0)(0,0) 278.842711 -4.179435 -4.113917 -4.152811
(1,2)(0,0) 279.995169 -4.166593 -4.057396 -4.122221
(2,0)(0,0) 278.843181 -4.164291 -4.076933 -4.128792
(1,1)(0,0) 278.843111 -4.164290 -4.076932 -4.128791
(0,2)(0,0) 278.548822 -4.159831 -4.072473 -4.124332
(0,1)(0,0) 277.415747 -4.157814 -4.092296 -4.131191
(4,0)(0,0) 280.357794 -4.156936 -4.025900 -4.103689
(0,3)(0,0) 279.078541 -4.152705 -4.043508 -4.108333
(3,0)(0,0) 278.923165 -4.150351 -4.041154 -4.105978
(0,4)(0,0) 279.408440 -4.142552 -4.011516 -4.089305
(0,0)(0,0) 267.671571 -4.025327 -3.981648 -4.007578

Model Selection Criteria Table


Dependent Variable: D(LFPRICESA)
Date: 05/10/18 Time: 11:58
Sample: 2007M01 2017M12
Included observations: 131
55

De La Salle University – Dasmariñas

Table 6 showed the ARIMA criteria table for D(LFPRICESA). As previously

shown in Table 5, there is no significant autocorrelation in both the AC and PAC plots.

The ARIMA terms, AR(p) and MA(q), are given the values 2 and 3, respectively. This

means that given the AIC, BIC, and HQ values, the ARIMA model chosen was (2,3)

(0,0). It can be concluded that the ARIMA Model (2,3)(0,0) is the best model in

forecasting the price of corn in the Philippines at first difference.

Step 6: Evaluation of the ARIMA Model

The ARIMA model is expressed as the form ARIMA (p,d,q), where p is the

autoregression order, d is the differencing level, and q is the moving average order.

There are four models that are chosen for comparison based on steps 4 to 5, satisfying

the following conditions, namely: (1) the ARMA model (3,3) using LFPRICESA at

level; (2) the ARMA model (3,3) at first difference; (3) the ARMA model (2,3) using

LFPRICESA at level; and (4) the ARMA model (2,3) using LFPRICESA at first

difference. Figures 3 to 6 show the evaluation of the ARIMA models (3,0,3), (3,1,3),

(2,0,3), and (2,1,3), respectively. The values of RMSE, MAE, MAPE, and AIC were

checked to compare the performances of the four models. The model that displays the

lowest value of the four criteria will be the best ARIMA model. After estimating the

models individually, the next step is proceed directly to forecasting, where the values of
56

De La Salle University – Dasmariñas

RMSE, MAE, and MAPE are shown. Tables 4 and 6 show the AIC values of the four

models. The results show that ARIMA model (3,1,3) is the best ARIMA model for

comparison against AR model in forecasting the price of corn in the Philippines on the

bases of the four mentioned values.


57

De La Salle University – Dasmariñas

2.7
Forecast: A
Actual: LFPRICESA
2.6
Forecast sample: 2007M01 2017M12
Adjusted sample: 2007M03 2017M12
2.5 Included observations: 130
Root Mean Squared Error 0.084588
2.4 Mean Absolute Error 0.067504
Mean Abs. Percent Error 2.757657
2.3
Theil Inequality Coefficient 0.017317
Bias Proportion 0.042072
Variance Proportion 0.522860
2.2
Covariance Proportion 0.435068

2.1
07 08 09 10 11 12 13 14 15 16 17

A ± 2 S.E.

Figure 3. Forecast evaluation graph ARIMA model (2,0,3)


58

De La Salle University – Dasmariñas

.08
Forecast: B
.06 Actual: D(LFPRICESA)
Forecast sample: 2007M01 2017M12
.04
Adjusted sample: 2007M04 2017M12
.02
Included observations: 129
Root Mean Squared Error 0.029642
.00 Mean Absolute Error 0.022025
Mean Abs. Percent Error 140.2943
-.02 Theil Inequality Coefficient 0.806206
Bias Proportion 0.000151
-.04
Variance Proportion 0.797569
-.06 Covariance Proportion 0.202280

-.08
07 08 09 10 11 12 13 14 15 16 17

B ± 2 S.E.

Figure 4. Forecast evaluation graph ARIMA model (2,1,3)


59

De La Salle University – Dasmariñas

2.7
Forecast: CC
Actual: LFPRICESA
2.6
Forecast sample: 2007M01 2017M12
Adjusted sample: 2007M04 2017M12
2.5 Included observations: 129
Root Mean Squared Error 0.080513
2.4 Mean Absolute Error 0.064124
Mean Abs. Percent Error 2.598020
2.3
Theil Inequality Coefficient 0.016524
Bias Proportion 0.161954
Variance Proportion 0.537861
2.2
Covariance Proportion 0.300185

2.1
07 08 09 10 11 12 13 14 15 16 17

CC ± 2 S.E.

Figure 5. Forecast evaluation graph ARIMA model (3,0,3)


60

De La Salle University – Dasmariñas

.08
Forecast: DD
.06 Actual: D(LFPRICESA)
Forecast sample: 2007M01 2017M12
.04
Adjusted sample: 2007M05 2017M12
.02
Included observations: 128
Root Mean Squared Error 0.029114
.00 Mean Absolute Error 0.021480
Mean Abs. Percent Error 148.0214
-.02 Theil Inequality Coefficient 0.772748
Bias Proportion 0.000330
-.04
Variance Proportion 0.757820
-.06 Covariance Proportion 0.241850

-.08
07 08 09 10 11 12 13 14 15 16 17

DD ± 2 S.E.

Figure 6. Forecast evaluation graph ARIMA model (3,1,3)


61

De La Salle University – Dasmariñas

Autoregressive Model
Table 7

Autoregression Estimates

LFPRICESA

LFPRICESA(-1) 1.329899
(0.07990)
[ 16.6437]

LFPRICESA(-2) -0.425318
(0.07803)
[-5.45061]

C 0.234847
(0.06597)
[ 3.55966]

R-squared 0.904805
Adj. R-squared 0.903306
Sum sq. resids 0.098570
S.E. equation 0.027859
F-statistic 603.5534
Log likelihood 282.5321
Akaike AIC -4.300495
Schwarz SC -4.234321
Mean dependent 2.450104
S.D. dependent 0.089592

Vector Autoregression Estimates


Date: 05/10/18 Time: 12:51
Sample (adjusted): 2007M03 2017M12
Included observations: 130 after adjustments
Standard errors in ( ) & t-statistics in [ ]
62

De La Salle University – Dasmariñas

The Autoregression estimates above showed the summary of the estimated AR

model. The right-hand side of the variable, LFPRICESA, shows the values of

coefficient, standard errors, and t-statistics. From here, there are four things that can be

concluded, namely: (1) the coefficient value of LFPRICESA at lag one of 1.329899

means that an increase of one percent in LFPRICESA(-1) will lead to an increase in the

value of LFPRICESA by 1.329899%; (2) likewise, the coefficient value of

LFPRICESA at lag two of -0.425318 means that as LFPRICESA(-2) increases by one

percent, LFPRICESA decreases by 0.425318%; (3) the standard errors calculate the

statistical reliability of the estimated coefficients, which means that the larger the value

of the standard errors, the greater the problem is observed in the estimates; and (4) to

interpret the t-statistic values, which is the coefficient value divided by the standard

error value, it is important to take a look at the value of probability of t-statistics.

With 2 lags set as the optimum number of lags, an AR equation has been

estimated using LFPRICESA. There are 2 independent variables, 1 independent

variable, with 3 coefficients in this model. So far, judging from the R-squared and F-

statistic, the model’s overall performance is good. However, as mentioned above, not

enough information is shown whether the independent values are significant enough to

explain the dependent variable.


63

De La Salle University – Dasmariñas

Table 8

Ordinary Least Squares Method

Coefficient Std. Error t-Statistic Prob.

C(1) 1.329899 0.079904 16.64375 0.0000


C(2) -0.425318 0.078031 -5.450610 0.0000
C(3) 0.234847 0.065975 3.559656 0.0005

Determinant residual covariance 0.000758

Equation: LFPRICESA = C(1)*LFPRICESA(-1) + C(2)*LFPRICESA(-2)


+ C(3)
Observations: 130
R-squared 0.904805 Mean dependent var 2.450104
Adjusted R-squared 0.903306 S.D. dependent var 0.089592
S.E. of regression 0.027859 Sum squared resid 0.098570
Durbin-Watson stat 2.049812

System: UNTITLED
64

De La Salle University – Dasmariñas

Estimation Method: Least Squares


Date: 05/10/18 Time: 12:58
Sample: 2007M03 2017M12
Included observations: 130
Total system (balanced) observations 130

Table 8 showed the OLS estimation of LFPRICESA after its AR estimation, as

shown by the 2 lags set. Even if Table 5 has shown that the overall performance is

good, it will still not be enough is the independent variables are insignificant. To check

whether the independent variables are significant or not, we need to determine the

probability of t-statistics first. If the value is less than 5%, then the variable is

significant, otherwise it is insignificant. The table above shows the coefficients of the

OLS model as well as their respective t-statistics probability values. The equation

written in the lower portion of the OLS estimation summary shows the corresponding

independent variables the coefficients are attributed into. The value of t-statistics

probability of LFPRICESA(-1) is found to be less than 5%, which implies its statistical

significance. In the same manner, the probability of t-statistics value of LFPRICESA(-

2) of 0.00% shows that it is statistically significant at 5%. Overall, both LFPRICESA(-

1) and LFPRICESA(-2) are significant enough to explain the changes in LFPRICESA.

The model proves that it is not spurious because the R-squared value is less than the

value of Durbin-Watson statistics. Therefore, the farmers can rely on this model when

making their decisions whether or not to produce more corn in the future.

Table 9
65

De La Salle University – Dasmariñas

Forecast Evaluation (AR Model)

Variable Inc. obs. RMSE MAE MAPE Theil

LFPRICESA 132 0.081858 0.064553 2.628879 0.016690

RMSE: Root Mean Square Error


MAE: Mean Absolute Error
MAPE: Mean Absolute Percentage Error

Forecast Evaluation
Date: 05/10/18 Time: 13:00
Sample: 2007M01 2017M12
Included observations: 132

Figures 3 to 6 evaluated the performance of the top 4 ARIMA models, and

selected among the best of them, while Table 9 showed the evaluation of the AR model.

When using the Root Mean Squared Error, the lower the value, the better the

performance of the model. In the case of Theil’s Inequality Coefficient, if the value is 1,

the forecasting model is called perfectly fit, which means that the actual and forecasted

values are the same. If the value is 0, then the predictive power of the forecasting model

is at its worst. Given that those 2 values are almost never seen in real life situation,

there exist values in between 0 and 1. The closer the value is to 1, the better the

performance of the forecasting model. On the basis of MAPE, the smaller the value is,

the lesser the forecast error becomes in percentage terms. The ARIMA model (3,1,3)

has an RMSE value of 0.029114, while the AR model has 0.081858. On the basis of

RMSE, clearly, the ARIMA model performed better. On Theil’s Inequality Coefficient,
66

De La Salle University – Dasmariñas

the ARIMA model (3,1,3) performed better given its value of 0.772748, as opposed to

the AR model’s value of 0.016690. Finally, the MAPE value of the ARIMA (3,1,3)

model is 14,802.14%, while the AR model’s MAPE value is 262.8879%. The ARIMA

model (3,1,3) failed proved to perform better than the AR model on the basis of MAPE.

Therefore, on the basis of RMSE, and Theil’s Inequality Coefficient values, the best

model to use in forecasting the farmgate prices of corn in the Philippines is the ARIMA

(3,1,3) Model.
67

De La Salle University – Dasmariñas

2.7

2.6

2.5

2.4

2.3

2.2

2.1
07 08 09 10 11 12 13 14 15 16 17 18

LFPRICESA_ARIMA LFPRICESA

Figure 7. Actual and forecasted prices using ARIMA

Figure 7 showed the graphical representation of the actual monthly prices from

2007 to 2017, as well as the forecasted prices on 2018, using ARIMA (3,1,3) model.

This graph shows that in 2018 will start with a few months of increasing prices, which

will be followed by some months of decreasing prices. After the months of decreasing

prices, the remaining months will be coupled with increasing prices. Overall, it is safe
68

De La Salle University – Dasmariñas

to assume that the corn farmers will earn profit if they are to start selling corn at

present.

Table 10

Actual and Forecasted Farmgate Prices of Corn Using ARIMA (3,1,3) Model

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
2007 January 2.21 -
February 2.22 -
March 2.27 -
69

De La Salle University – Dasmariñas

April 2.31 -
May 2.34 -
June 2.34 -
July 2.32 -
August 2.33 -
September 2.34 -
October 2.34 -
November 2.32 -
December 2.31 -
2008 January 2.31 -
February 2.32 -
March 2.35 -
April 2.38 -
May 2.35 -
June 2.38 -
July 2.37 -
August 2.37 -
September 2.41 -
October 2.43 -
November 2.45 -
December 2.47 -
2009 January 2.54 -
February 2.57 -
March 2.48 -
April 2.47 -
May 2.42 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
June 2.32 -
July 2.30 -
August 2.22 -
September 2.20 -
70

De La Salle University – Dasmariñas

October 2.21 -
November 2.28 -
December 2.28 -
2010 January 2.30 -
February 2.34 -
March 2.40 -
April 2.40 -
May 2.38 -
June 2.42 -
July 2.42 -
August 2.42 -
September 2.40 -
October 2.40 -
November 2.44 -
December 2.49 -
2011 January 2.51 -
February 2.51 -
March 2.50 -
April 2.44 -
May 2.43 -
June 2.43 -
July 2.42 -
August 2.41 -
September 2.43 -
October 2.50 -
November 2.57 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
December 2.57 -
2012 January 2.59 -
February 2.57 -
March 2.50 -
71

De La Salle University – Dasmariñas

April 2.50 -
May 2.52 -
June 2.51 -
July 2.51 -
August 2.52 -
September 2.55 -
October 2.56 -
November 2.56 -
December 2.54 -
2013 January 2.50 -
February 2.44 -
March 2.44 -
April 2.45 -
May 2.46 -
June 2.47 -
July 2.46 -
August 2.47 -
September 2.52 -
October 2.52 -
November 2.49 -
December 2.48 -
2014 January 2.46 -
February 2.44 -
March 2.47 -
April 2.50 -
May 2.54 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
June 2.58 -
July 2.58 -
August 2.66 -
September 2.62 -
72

De La Salle University – Dasmariñas

October 2.60 -
November 2.55 -
December 2.53 -
2015 January 2.49 -
February 2.48 -
March 2.51 -
April 2.51 -
May 2.52 -
June 2.52 -
July 2.53 -
August 2.54 -
September 2.53 -
October 2.50 -
November 2.47 -
December 2.45 -
2016 January 2.44 -
February 2.47 -
March 2.50 -
April 2.50 -
May 2.53 -
June 2.55 -
July 2.56 -
August 2.55 -
September 2.51 -
October 2.44 -
November 2.41 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
December 2.40 -
2017 January 2.40 -
February 2.40 -
March 2.40 -
73

De La Salle University – Dasmariñas

April 2.41 -
May 2.42 -
June 2.41 -
July 2.41 -
August 2.44 -
September 2.50 -
October 2.50 -
November 2.52 -
December 2.52 -
2018 January - 2.54
February - 2.55
March - 2.55
April - 2.55
May - 2.53
June - 2.52
July - 2.51
August - 2.51
September - 2.51
October - 2.53
November - 2.55
December - 2.56
74

De La Salle University – Dasmariñas

2.7

2.6

2.5

2.4

2.3

2.2

2.1
07 08 09 10 11 12 13 14 15 16 17 18

LFPRICESA_AR LFPRICESA

Figure 8. Actual and forecasted prices using AR

The graph above showed the actual and forecasted prices using the AR model.

Contrary to what is shown by the results of using ARIMA model, the AR model

predicted that prices will continuously increase throughout the entire 2018. This only

means that if the corn farmers were to start selling corn at present, they will earn profit

by the end of 2018. This graphical result, as far as earning profit is concerned, agrees
75

De La Salle University – Dasmariñas

that ARIMA (3,1,3) model is a better model in forecasting the price of corn in the

Philippines.

Table 11

Actual and Forecasted Farmgate Prices of Corn Using AR Model

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
2007 January 2.21 -
February 2.22 -
March 2.27 -
76

De La Salle University – Dasmariñas

April 2.31 -
May 2.34 -
June 2.34 -
July 2.32 -
August 2.33 -
September 2.34 -
October 2.34 -
November 2.32 -
December 2.31 -
2008 January 2.31 -
February 2.32 -
March 2.35 -
April 2.38 -
May 2.35 -
June 2.38 -
July 2.37 -
August 2.37 -
September 2.41 -
October 2.43 -
November 2.45 -
December 2.47 -
2009 January 2.54 -
February 2.57 -
March 2.48 -
April 2.47 -
May 2.42 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
June 2.32 -
July 2.30 -
August 2.22 -
September 2.20 -
77

De La Salle University – Dasmariñas

October 2.21 -
November 2.28 -
December 2.28 -
2010 January 2.30 -
February 2.34 -
March 2.40 -
April 2.40 -
May 2.38 -
June 2.42 -
July 2.42 -
August 2.42 -
September 2.40 -
October 2.40 -
November 2.44 -
December 2.49 -
2011 January 2.51 -
February 2.51 -
March 2.50 -
April 2.44 -
May 2.43 -
June 2.43 -
July 2.42 -
August 2.41 -
September 2.43 -
October 2.50 -
November 2.57 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
December 2.57 -
2012 January 2.59 -
February 2.57 -
March 2.50 -
78

De La Salle University – Dasmariñas

April 2.50 -
May 2.52 -
June 2.51 -
July 2.51 -
August 2.52 -
September 2.55 -
October 2.56 -
November 2.56 -
December 2.54 -
2013 January 2.50 -
February 2.44 -
March 2.44 -
April 2.45 -
May 2.46 -
June 2.47 -
July 2.46 -
August 2.47 -
September 2.52 -
October 2.52 -
November 2.49 -
December 2.48 -
2014 January 2.46 -
February 2.44 -
March 2.47 -
April 2.50 -
May 2.54 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
June 2.58 -
July 2.58 -
August 2.66 -
September 2.62 -
79

De La Salle University – Dasmariñas

October 2.60 -
November 2.55 -
December 2.53 -
2015 January 2.49 -
February 2.48 -
March 2.51 -
April 2.51 -
May 2.52 -
June 2.52 -
July 2.53 -
August 2.54 -
September 2.53 -
October 2.50 -
November 2.47 -
December 2.45 -
2016 January 2.44 -
February 2.47 -
March 2.50 -
April 2.50 -
May 2.53 -
June 2.55 -
July 2.56 -
August 2.55 -
September 2.51 -
October 2.44 -
November 2.41 -
Continued

Year Month Actual Price Forecasted Price


(in PhP/kg.) (in PhP/kg.)
December 2.40 -
2017 January 2.40 -
February 2.40 -
March 2.40 -
80

De La Salle University – Dasmariñas

April 2.41 -
May 2.42 -
June 2.41 -
July 2.41 -
August 2.44 -
September 2.50 -
October 2.50 -
November 2.52 -
December 2.52 -
2018 January - 2.52
February - 2.53
March - 2.53
April - 2.53
May - 2.53
June - 2.53
July - 2.54
August - 2.54
September - 2.54
October - 2.54
November - 2.54
December - 2.55

Differences Between the Actual Data and Predictive Data

This last part of the chapter discussed about the gap between the actual data and

the predictive data using AR and ARIMA (3,1,3) models. The previous discussions

showed how to estimate the forecasting models ARIMA (3,1,3) and AR, and how

accurate these two models are in forecasting future prices. Provided in this section is a

discussion on which model has the closer predictive data to the actual data.
81

De La Salle University – Dasmariñas

The results indicated that there is a huge gap between the predictive data using

ARIMA (3,1,3) model and the actual data. While the de-seasonalized logarithmic data

had a lowest value of 2.20 and a highest value of 2.66, the predictive value revolved

around 0.0. This is not a good indication for the farmers to plant corn if they are to

adapt this model in their production decisions. On the other hand, there is a

considerable gap between the predictive data using AR model and the actual data. The

predictive value had a starting value of 2.29 in 2007, and 2.52 in 2017. The farmers will

most likely earn profit from producing corn at present, ceteris paribus.

Table 10

ARIMA Model (3,1,3) OLS Estimation

Variable Coefficient Std. Error t-Statistic Prob.

C 0.002216 0.003366 0.658180 0.5116


AR(1) 1.648502 0.023556 69.98140 0.0000
AR(2) -0.983811 0.023713 -41.48812 0.0000
MA(1) -1.396783 0.096188 -14.52142 0.0000
MA(2) 0.658443 0.137901 4.774740 0.0000
82

De La Salle University – Dasmariñas

MA(3) 0.186109 0.090752 2.050744 0.0424


SIGMASQ 0.000688 8.31E-05 8.285646 0.0000

R-squared 0.300043 Mean dependent var 0.002399


Adjusted R-squared 0.266174 S.D. dependent var 0.031480
S.E. of regression 0.026967 Akaike info criterion -4.316214
Sum squared resid 0.090175 Schwarz criterion -4.162578
Log likelihood 289.7120 Hannan-Quinn criter. -4.253785
F-statistic 8.858944 Durbin-Watson stat 2.003855
Prob(F-statistic) 0.000000

Inverted AR Roots .82-.55i .82+.55i


Inverted MA Roots .79-.58i .79+.58i -.19

Dependent Variable: D(LFPRICESA)


Method: ARMA Maximum Likelihood (OPG - BHHH)
Date: 05/10/18 Time: 13:08
Sample: 2007M02 2017M12
Included observations: 131
Convergence achieved after 44 iterations
Coefficient covariance computed using outer product of gradients

Table 10 showed the OLS estimation of ARIMA model with AR(3), I(1) and

MA(3) terms. This is the exact ARIMA model used to check the gap between the actual

and predictive prices. The R-squared is the measurement of how successful the

regression model is. In this model, it means the fraction of the variance of

D(LFPRICESA) that can be explained by the independent variables. If it shows a value

of 100%, it means that the regression fits perfectly, which is not practical in real life

situations. In this model the R-squared value is 30.0043%, and this means that the

independent variables in this model explain 30.0043% of the changes in the dependent

variable, D(LFPRICESA).
83

De La Salle University – Dasmariñas

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5
07 08 09 10 11 12 13 14 15 16 17

LFPRICESA LFPRICESA_ARIMA

Figure 9. Forecast graph


84

De La Salle University – Dasmariñas

It can be seen from the graph the comparison between the actual and forecasted

values of LFPRICESA. There exists a considerable gap between the actual value and

the forecasted value. The actual values exhibited a nearly horizontal trend, with few

price fluctuations. On the contrary, the forecasted values exhibited a somewhat volatile,

yet neither increasing nor decreasing, trend. Overall, the assumption is that in the long-

run, due to the forecasted prices being low, it is not advisable for the corn farmers to

produce corn at present as they will only be suffering huge losses in doing so.
85

De La Salle University – Dasmariñas

2.7

2.6

2.5

2.4

2.3

2.2

2.1
07 08 09 10 11 12 13 14 15 16 17

LFPRICESA LFPRICESA_AR

Figure 10. Forecast graph


86

De La Salle University – Dasmariñas

The graph above showed that both the actual and forecasted values exhibit

upward trends. The obvious difference is that the actual values have a lot of price

fluctuations, while the forecasted values showed a steady increasing trend at an

increasing rate. This only suggests that in the long-run, with the right timing, the

farmers somehow might be able to enjoy earning a profit from producing corn at

present. However, it can be observed that the profit the farmers will gain in the future

will not be enough to cover the losses that they will most likely incur.
87

De La Salle University – Dasmariñas

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5
07 08 09 10 11 12 13 14 15 16 17

LFPRICESA
LFPRICESA_AR
LFPRICESA_ARIMA

Figure 11. Forecast comparison graph


88

De La Salle University – Dasmariñas

Figure 11 presented the graphical comparison summary of the performance

models of ARIMA and AR models. In the graphs presented above, it was proven that

the AR model would most likely be more helpful to the corn farmers as opposed the

ARIMA model. However, the AR model only suggests that the farmers will be able to

earn much profit if they were to rely on this model in their production decisions. If

there are any other models that will give the farmers the opportunity to earn more and

lose less, then it will be more strongly preferred.

CHAPTER VI
89

De La Salle University – Dasmariñas

SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

Summary

The research was primarily concerned in determining the model that would best

forecast the farmgate prices of corn in the Philippines. The researcher gathered the

monthly farmgate corn prices directly from the PSA. Given the wide range of

forecasting models available, the researcher decided to compare the performances of

ARIMA and AR models in this univariate analysis. The main motivation for this study

is the fact that the corn is second most important crop in the Philippines and almost no

studies were made regarding forecasting the price of the said commodity. Without the

imported goods from abroad, the Filipinos would turn to the appetizers such as Filipino

bread, corn, and mashed potatoes. Of course, the decrease in demand for corns will

strongly affect the producers. Thus, this study was made to somehow help the

producers in their future decisions. Both historical and evaluative methods were used in

this study, which covered the years 2007 to 2017.

Tables 1 showed the trend of the farmgate prices of corn in the Philippines from

2007 to 2017. It was found out that regardless of the varying price fluctuations, the

farmgate prices exhibit a continuous upward trend, which led to an assumption that in

the long-run, prices of corn will further increase given various factors.

Before estimating the equations, it is necessary first to check whether the overall

model performs good and whether the variables are significant or not. In the case of

ARIMA model, there were six steps followed in order to provide a good ARIMA
90

De La Salle University – Dasmariñas

model. Overall, it was needed to convert the prices into logarithmic form, and so far

there was no need to difference the dependent variable. After the ARIMA model, the

AR model was estimated first, and looking at the R-squared and F-statistic, the overall

model is good. Next, an OLS regression is estimated to find out whether the

independent variables, LFPRICESA(-1) and LFPRICESA(-2) are significant enough

individually to explain the changes in the dependent variable, LFPRICE, and it turned

out that both LFPRICESA(-1) and LFPRICESA(-2) are statistically significant enough

to explain the changes in LFPRICE given the probability of t-statistics values. From

there, a forecasting model based on AR can be estimated.

Conclusions

The univariate study was conducted in order to compare the performances of

ARIMA and AR models, and to determine whether which of these models performs

better in forecasting the farmgate prices of corn in the Philippines. The results show

that given the values of R-squared and F-statistic, the AR performed well and that both

the independent variables are significant. The AR proved to be non-spurious because its

R-squared value exceeds its Durbin-Watson statistic value. The ARIMA (3,1,3) model,

however, had an R-squared value of 30.2949%, and this is most likely due to the lack of

independent variables accounting for many different factors that can possibly affect

D(LFPRICESA). The deciding factor of these two models is when their MAPE, RMSE,

and Theil’s Inequality Coefficient Values are checked. When looking into the RMSE,

when the value is lower, it means that the predictive capacity of a forecasting model is
91

De La Salle University – Dasmariñas

better. On the other hand, Theil’s Inequality Coefficient might exhibit 3 kinds of values,

namely: (1) 0, where the forecast model’s predictive power is at its worst; (2) 1

(perfectly fit), where the actual and forecasted values are the same; and (3) in between

0 and 1, wherein the predictive power of the forecasting model becomes better as the

value approaches near 1. Lastly, the lower the value of MAPE is, the lesser the forecast

error becomes. The results showed that the RMSE value of ARIMA Model is lower

than that of AR Model’s, the former model’s Theil’s Inequality Coefficient value is

closer to 1 than the latter model, and the latter model’s MAPE value is way lower than

that of the former model’s. In this study, satisfying two among the three criteria, the

ARIMA model proved to be the best model to use in forecasting the farmgate prices of

corn in the Philippines.

Recommendations

The research was able to show clearly that the ARIMA Model is better than the

AR Model. However, some things have to be taken into consideration. First, the

ARIMA model was only able to outperform the AR model in two out of three criteria.

This only suggests that not all cases will the ARIMA model perform better against the

AR model.

This is a good study which is not thoroughly looked up to in the Philippines.

Given that situation, it is advised for the future researchers to work on other models,

aside from ARIMA and AR, which can be used to compare performances with the two

models in this study.


92

De La Salle University – Dasmariñas

Next, there is only one variable used in this study, and that is the price of corn

itself. This means that this study assumed that corn prices can be assumed, ceteris

paribus. This also became a problem because the ARIMA (3,1,3) forecasting model had

a low R-squared value. So adding more independent variables will do much help. For

the future researchers, they are recommended to find other variables that might be

useful in forecasting corn prices. That will make this study much more realistic when

other factors will be included that will greatly make or break the performance of the

forecasting model.

This study is only limited to the corn prices in the Philippines. The future

researchers, and the government as well are recommended to gather the data of corn

prices, assuming they are available and accessible, to the neighboring ASEAN countries

for a more comprehensive study. That study will not only benefit the Filipino people,

but also those neighboring ASEAN countries.

REFERENCES

Adeyanju, C. (2014). The Top Factors that Move the Price of Corn. Retrieved from
https://www.futuresknowledge.com/news-and-analysis/grains/the-top-factors-
that-move-the-price-of-corn/.
93

De La Salle University – Dasmariñas

Ahmad, F. (2013). The Effect of Oil Prices on Unemployment: Evidence from Pakistan.
Retrieved from http://www.berjournal.com/wp-content/plugins/downloads-
manager/upload/BERJ%204(1)13%20Article%203%20pp.43-57.pdf.

Alquist, R. and Kilian, L. (2010). What do we Learn from the Price of Crude Oil
Futures? Retrieved from
https://deepblue.lib.umich.edu/bitstream/handle/2027.42/75776/1159_ftp.pdf?
sequence=1.

Alquist, R., Kilian, L. and Vigfusson, R. J. (2011). Forecasting the Price of Oil.
Retrieved from
https://www.federalreserve.gov/pubs/ifdp/2011/1022/ifdp1022.pdf.

Agricultural Management Assistance. (2008). Zea mays. Retrieved from


http://eol.org/pages/1115259/details.

Bäckman, S. and Sumelius, J. (2009). Identifying the Driving Forces Behind Price
Fluctuations and Potential Food Crisis. Retrieved from
http://www.helsinki.fi/taloustiede/Abs/DP35.pdf.

Bowman, C. and Husain, A. M. (2004). Forecasting Commodity Prices: Futures Versus


Judgment. Retrieved from
https://www.imf.org/external/pubs/ft/wp/2004/wp0441.pdf.

Chinn, M. D. and Coibion, O. (2010). The Predictive Content of Commodity Futures.


Retrieved from http://www.nber.org/papers/w15830.pdf.

Corpuz, P. G. (2010). Philippine Grain and Situation Outlook. Retrieved from


https://gain.fas.usda.gov/Recent%20GAIN%20Publications/Grain%20and
%20Feed%20Annual_Manila_Philippines_2-5-2010.pdf.

Dalinina, R. (2017). Introduction to Forecasting with ARIMA in R. Retrieved from


https://www.datascience.com/blog/introduction-to-forecasting-with-arima-in-r-
learn-data-science-tutorials.

Dash, M., Solanki, A. and Shobana, T. (2012). A Study on Commodity Market


Behaviour, Price Discovery and Its Factors. Retrieved from
https://www.researchgate.net/publication/228150427_A_Study_on_Commodity
_Market_Behaviour_Price_Discovery_and_Its_Factors.

Dönmez, A. and Magrini, E. (2013). Agricultural Commodity Price Volatility and Its
Macroeconomic Determinants. Retrieved from
http://ftp.jrc.es/EURdoc/JRC84138.pdf.
94

De La Salle University – Dasmariñas

Food and Agriculture Organization Corporate Statistical Database. (2012). Zea mays.
Retrieved from http://eol.org/pages/1115259/details.

Frankel, J. A. and Rose, A. K. (2009). Determinants of Agricultural and Mineral


Commodity Prices. Retrieved from
https://www.rba.gov.au/publications/confs/2009/frankel-rose.html.

GMA News Online. (2013). Yolanda Deals Central PHL P3.7B in Agri Damage,
Supplies Stable – DA. Retrieved from
http://www.gmanetwork.com/news/money/content/334930/yolanda-deals-
central-phl-p3-7b-in-agri-damage-supplies-stable-da/story/.

Good, D. (2008). Many Factors Influencing Corn and Soybean Prices. Retrieved from
http://www.farmdoc.illinois.edu/marketing/weekly/html/092208.html.

Gwirtz, J. A. and Casal, M. G. (2014). Processing Maize Flour and Corn Meal Food
Products. Retrieved from
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4260129/pdf/nyas1312-
0066.pdf.

Halonen, D. G. (2016). Forecasting the Spot Price of Corn: Methods and Assessment.
Retrieved from http://repository.stcloudstate.edu/cgi/viewcontent.cgi?
article=1006&context=econ_etds.

Hamjah, M. A. (2014). Forecasting Major Fruit Crops Productions in Bangladesh using


Box-Jenkins ARIMA Model. Retrieved from
http://iiste.org/Journals/index.php/JEDS/article/viewFile/12578/12897.

Hillier, F. S. and Hillier, M. S. (2001). Introduction to Operations Research. Retrieved


from https://notendur.hi.is/kth93/3.20.pdf.

Hyndman, R. J. (2014). Measuring Forecast Accuracy. Retrieved from


http://citeseerx.ist.psu.edu/viewdoc/download?
doi=10.1.1.434.2646&rep=rep1&type=pdf.

Hyndman, R. J. and Koehler, A. B. (2005). Another Look at Measures of Forecast


Accuracy. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?
doi=10.1.1.154.9771&rep=rep1&type=pdf.

Jha, G. K. and Sinha, K. (2013). Agricultural Price Forecasting Using Neural Network
Model: An Innovative Information Delivery System. Retrieved from
http://ageconsearch.umn.edu/bitstream/162150/2/8-GK-Jha.pdf.
95

De La Salle University – Dasmariñas

Jadhav, V., Reddy, B. V. and Gaddi, G. M. (2017). Application of ARIMA Model for
Forecasting Agricultural Prices. Retrieved from
http://jast.modares.ac.ir/article_16988_5dcee1eff8387db448a1ee6725b3aa72.pd
f.

Jorgensen, N. (2014). A Comparison of Corn Yield Forecasting Models. Retrieved from


http://openprairie.sdstate.edu/cgi/viewcontent.cgi?article=1339&context=etd.

Juliano, B. O. and Gonzales, L. A. (n.d.). Rice and Corn Technology and Policy in the
Philippines. Retrieved from http://www.nast.ph/images/pdf
%20files/Publications/NAST%20Transactions/NAST%201986%20Transactions
%20Volume%208/Plenary%20Paper%201%20Rice%20and%20Corn
%20Technology%20and%20Policy%20in%20the%20Philippines
%20%20%20Bienvenido%20O.%20Juliano%20and%20Leonardo%20A.
%20Gonzales.pdf.

Kulkarni, S. and Haidar, I. (2009). Forecasting Model for Crude Oil Price Using
Artificial Neural Networks and Commodity Futures Prices. Retrieved from
https://arxiv.org/ftp/arxiv/papers/0906/0906.4838.pdf.

Latham Hi-Tech Seeds. (2012). The Importance of Corn. Retrieved from


http://www.thefieldposition.com/2012/06/the-importance-of-corn/.

Li, G., Xu, S. and Li, Z. (2010). Short-Term Price Forecasting For Agro-products Using
Artificial Neural Networks. Retrieved from http://ac.els-
cdn.com/S2210784310000367/1-s2.0-S2210784310000367-main.pdf?
_tid=2aecff02-19c1-11e7-b21b-
00000aacb360&acdnat=1491370489_9c6a3b54b1b21f46a5e3fdd2d4427431.

Mckenzie, A. M. and Holt, M. T. (1998). Market Efficiency in Agricultural Futures


Markets. Retrieved from
https://ageconsearch.umn.edu/bitstream/20933/1/spmcke01.pdf.
96

De La Salle University – Dasmariñas

Off The Grid News. (2015). The Importance of Corn in The American Economy.
Retrieved from http://www.offthegridnews.com/off-grid-foods/the-importance-
of-corn-in-the-american-economy/.

Okunmadewa, F. (n.d.). Enhancing Farm Gate Prices: The Role of Agricultural


Marketing Research. Retrieved from
http://ageconsearch.umn.edu/bitstream/147523/2/Prof.%20Okumadewa.pdf.

Oster, A. (2015). Commodity Price Declines and Their Economic Impact. Retrieved
from https://internationalbanker.com/banking/commodity-price-declines-and-
their-economic-impact/.

Padin, M. G. (2016). El Niño Causes Spike in Local Corn Prices. Retrieved from
https://businessmirror.com.ph/el-nino-causes-spike-in-local-corn-prices/.

Ribeiro, C. O. and Oliveira, S. M. (2011). A Hybrid Commodity Price-forecasting


Model Applied to the Sugar–alcohol Sector. Retrieved from
http://onlinelibrary.wiley.com/doi/10.1111/j.1467-8489.2011.00534.x/full.

Sanders, D. J. and Baker, T. G. (2012). Forecasting Corn and Soybean Basis Using
Regime-Switching Models. Retrieved from
http://www.farmdoc.illinois.edu/nccc134/conf_2012/pdf/confp16-12.pdf.

Sands, S. (2015). How the World Economy Adjusts to Lower Commodity Prices.
Retrieved from http://marketrealist.com/2015/10/world-adjusts-lower-
commodity-prices/.

Sims, C. A. (1980). Macroeconomics and Reality. Retrieved from


http://www.ekonometria.wne.uw.edu.pl/uploads/Main/macroeconomics_and_re
ality.pdf.

Taleghani, S., Aslani, S. and Shiry, S. (2008). Robust Moving Object Detection from a
Moving Video Camera Using Neural Network and Kalman Filter. Retrieved
from
https://www.researchgate.net/publication/220797913_Robust_Moving_Object_
Detection_from_a_Moving_Video_Camera_Using_Neural_Network_and_Kalm
an_Filter.

Teucrium Trading, LLC. (2015). Seasonal Price Patterns in Corn: December Might be
the Best Month to Consider Investing in Corn. Retrieved from
97

De La Salle University – Dasmariñas

https://seekingalpha.com/article/3727656-seasonal-price-patterns-corn-
december-might-best-month-consider-investing-corn.

Tiffany, D. G. (2009). Economic and Environmental Impacts of U.S. Corn Ethanol


Production and Use. Retrieved from
https://files.stlouisfed.org/files/htdocs/publications/red/2009/01/Tiffany.pdf.

Timmer, C. P. (2008). Causes of High Food Prices. Retrieved from


https://www.adb.org/sites/default/files/publication/28375/economics-wp128.pdf.

Tripathi, R., Nayak, A. K., Raja, R., Shahid, M., Kumar, A., Mohanty, S. et al. (2014).
Forecasting Rice Productivity and Production of Odisha, India, Using
Autoregressive Integrated Moving Average Models. Retrieved from
https://www.hindawi.com/journals/aag/2014/621313/.
Ubilava, D. and Helmers, C. G. (2011). The ENSO Impact on Predicting World Cocoa
Prices. Retrieved from
http://ageconsearch.umn.edu/bitstream/103528/2/Ubilava_Helmers_2011.pdf.

United States Department of Agriculture Economic Research Service. (2017). Trade.


Retrieved from https://www.ers.usda.gov/topics/crops/corn/trade.aspx#world.

United States Congress Joint Economic Committee. (2013). The Economic


Contribution of America’s Farmers and the Importance of Agricultural Exports.
Retrieved from https://www.jec.senate.gov/public/_cache/files/266a0bf3-5142-
4545-b806-ef9fd78b9c2f/jec-agriculture-report.pdf.

Wisner, R. (2014). Ethanol, Gasoline, Crude Oil and Corn Prices: Are the Relationships
Changing?. Retrieved from https://www.agmrc.org/renewable-
energy/ethanol/ethanol-gasoline-crude-oil-and-corn-prices-are-the-relationships-
changing/.

S-ar putea să vă placă și