Essays on Volatility in the Crude Oil and Natural Gas Markets: GARCH, Asymmetry, Seasonality and Announcement Effects

Duong T. Le* Division of Finance Price College of Business University of Oklahoma Norman, OK 73019-0450 Ph: (405)325-3486 le@ou.edu This version: 5/12/2008

Abstract These essays examine the cause and behavior of actual and implied volatility in the US crude oil and natural gas markets. Although crude oil and natural gas prices are among the most volatile, they have received little academic scrutiny heretofore. I theorize and find that (1) the crude oil and natural gas markets are characterized by volatility persistence, (2) crude oil and natural gas implied volatilities are higher on shorter term options, (3) there is a skew pattern in crude oil and natural gas implied volatilities (4) negative and positive shocks have different impacts on both predicted volatility and implied volatility in these markets, (5) there are time-of-the-year and day-of-the-week effects on both actual and implied volatilities, (6) the actual volatilities increase and the implied volatilities decrease following relevant scheduled announcements and (7) these announcements have no further impact on volatility on subsequent days. I develop and employ an improved procedure to test for the determinants of volatility within GARCH type model.

* I am grateful to Louis Ederington for helpful comments and suggestions. I benefited from the comments of Duane Stock, Chitru Fernando and participants in the Finance workshop at the University of Oklahoma. I also appreciate comments received from Marian Turac and seminar participants at the 2008 Eastern Finance Association annual meeting. I thank the Price College of Business at the University of Oklahoma for financial support. All mistakes remain my own responsibility.

This dissertation consists of two parts. The first part explores the cause and behavior of price volatility in the crude oil and natural gas markets. The second part investigates the attributes of implied volatilities on options traded in the crude oil and natural gas markets. I. Introduction

Crude oil and natural gas are two of the most essential energy sources in the U.S., accounting for about 40% and 22% of the nations energy consumption, respectively. Since OPECs 1973 decision to regulate its oil price independently of large oil companies, crude oil prices have been subject to dramatic volatility, especially in recent years. From less than $11 per barrel in the beginning of 1999, oil prices increased to $38 per barrel in September 2000, decreased to $18 per barrel in January 2002, going up to $77 per barrel in July 2006, sliding down to $51 per barrel in January 2007 before reaching an unprecedented level of $125 per barrel in May 2008. Natural gas is also one of the most volatile markets, particularly since its evolution from a highly regulated market to a largely deregulated market in which prices are driven by supply and demand. In 2007, the annualized standard deviation of the daily percentage change in prices was 31.33% for crude oil and 50.23% for natural gas. By comparison, that number was only 4.08% for the US dollar-Euro exchange rate, 16.37% for the S&P 500 and 19.10% for the 10-year T-bond interest rates1. The high volatility in crude oil prices is likely due to supply uncertainty and the short term inelasticity of demand. Given that crude oil is the most essential energy for both industrial and residential sectors, it is very difficult for oil users to reduce their consumption within a short period of time. On the other hand, there is considerable uncertainty in oil supply which depends on a variety of macroeconomic and political factors. For example, in 1997, when the world economy was already in a recession, major oil producing countries, failing to predict the oil demand correctly, increased their production levels which resulted in a huge decrease in oil prices. Similarly, the high volatility in natural gas prices is likely due to the short-term inelasticity of supply and demand. Natural gas supplies are often constrained by storage levels and thus, natural gas suppliers are unable to increase production levels in a short period of time. Also, it is difficult for consumers to quickly reduce their consumption even when a sharp increase in natural gas prices occurs, especially during the winter. Since natural gas suppliers cannot rapidly adjust their production levels to match demand changes, supply and demand imbalances may result in sharp price changes. This high variability in crude oil and natural gas prices makes it extremely difficult for consumers to forecast their costs and for producers to forecast their profits. The desire to protect market participants against such price fluctuations has led to the creation of and active trading in oil and gas futures, swaps and options where the market value of the latter depends on volatility.

The data for the US dollar-Euro exchange rate, S&P 500 and the 10-year T-bond interest rates were collected from http://www.oanda.com, CRSP and the Federal Reserve website (http://www.federalreserve.gov), respectively.

While volatility in other markets has been a subject of intense research activity in recent decades, little attention has been paid to volatility in energy prices. Despite the fact that crude oil and natural gas prices tend to be more volatile than most other prices, research into the cause and behavior of crude oil and natural gas volatilities is limited. For instance, in a well-known and comprehensive study of the volatility literature, Poon and Granger (2003) surveyed 93 articles examining volatility in all sorts of markets; only three of these include crude oil among the markets examined (Day and Lewis (1993), Szakmary, Ors and Kim (2003) and Martens and Zein (2004)) and only Szakmary et al. (2003) included natural gas volatility. Among the limited studies on crude oil volatility not surveyed by Poon and Granger (2003) are those of Wilson, Aggarwal and Inclan (1996), Yang, Hwang and Huang (2002), Pindyck (2004), and Kuper and Soest (2006) who found that volatility clustering is an attribute of the crude oil market. However, none of the above papers examined other attributes of crude oil volatility such as asymmetric volatility, seasonality or announcement effects. Among the few studies on natural gas volatility are those of Susmel and Thompson (1997) who found that natural gas volatility follows an ARCH type process, of Szakmary et al. (2003) who documented the information content of implied volatility on natural gas options and of Linn and Zhu (2004) who documented that natural gas volatility increases on days the Weekly Natural Gas Storage Report is announced. An understanding of the cause and behavior of volatility in the crude oil and natural gas markets is essential to market participants since the market value of risk management products such as options and swaps depends mostly on volatility. Similarly, the only unknown factor in option pricing is volatility of the underlying asset. II. Essay 1: Volatility in the Crude oil and Natural gas markets: GARCH, Asymmetry, Seasonality and Announcement effects In the first part of my dissertation, I examine volatility in the crude oil and natural gas markets and attempt to answer the following questions: 1. Are crude oil and natural gas prices characterized by volatility persistence as has been documented in other markets? It has been observed that in many other markets volatile periods tend to follow volatile periods whereas stable periods tend to follow stable periods. Among numerous studies documenting volatility persistence are: Adrian, Pagan and Schwert (1990), Andersen, Bollerslev, Diebold and Ebens (2001), Wu (2001) and Flannery and Protopapadakis (2002) for the stock market, Ederington and Lee (1993, 1995 and 2001) for interest rates, Harvey and Huang (1991), Ederington and Lee (1993, 1995 and 2001), Andersen and Bollerslev (1998) and Low and Zhang (2005) for the foreign exchange market and Jones et al. (1998) for the Treasury bond market. I hypothesize that similar volatility persistence exists in the crude oil and natural gas markets. 2

2. Is there volatility asymmetry in the crude oil and natural gas markets? That is, do equal positive and negative shocks have different impacts on future volatility? This hypothesis is inspired by the generally documented evidence of an asymmetric volatility in the stock market and, to a lesser extent, in some other markets such as the Treasury bond and Treasury bond futures markets. French and Roll (1986), French, Schwert and Stambaugh (1987), Campbell and Hentschel (1992), Veronesi (1999), Bekaert and Wu (2000) and Wu (2001) found that in the stock market, an unexpected decrease in price has a bigger impact on predicted volatility than an unexpected increase in price of equal magnitude. Contrary to the evidence from the stock market, I think there are good reasons to expect a negative shock in the oil and gas markets to have a smaller impact on predicted volatility than an equivalent positive shock. My reasoning behind this hypothesis is the likely shape of the oil and gas supply and demand curves. At low volume and prices, oil and gas supply is highly elastic, but once storage limits are reached, supply becomes quite inelastic as oil and gas producers, due to infrastructure constraints, cannot increase their production levels within a short period of time. The demand curve for oil and gas also consists of an elastic portion when prices are low and an inelastic portion when prices are high. Given the hypothesized shape of the supply and demand curves, the same fluctuation in demand when prices are low should cause a smaller change in prices than when prices are high. Thus, a positive price shock which moves the oil and gas markets up the supply and demand curves is likely to presage higher future volatility than a negative shock moving the markets down the curves. 3. Does volatility differ by day of the week? Some academic studies found that volatility of asset returns varies across days of the week2 and that Monday volatility (including weekend) is higher than volatility of a normal one-day period but not as high as that of a three-weekday period presumably because there is not much information coming out during the weekend. Consistent with the evidence from other markets, I expect that crude oil and natural gas volatilities differ across days of the week and increase over the weekend. Furthermore, I hypothesize that the higher weekend volatility is particularly likely for natural gas since natural gas price fluctuations supposedly depend on weather conditions which are just as likely to change on weekend days as on weekdays. To my knowledge, no one has looked at these questions. 4. Is there seasonality in crude oil and natural gas volatilities? A winter effect is likely unique in the natural gas market due to the dependence of natural gas prices on weather conditions. Natural gas prices are among the most sensitive to weather conditions as the U.S. typically consumes twice as much natural gas in the winter as in the summer (due to space heating). At the same time, as the U.S. natural gas production is relatively constant throughout the year (due to storage capacity) and overseas imports are very limited, the supply of natural gas in the winter is essentially fixed. Therefore, supply and demand imbalances in the natural
The literature on day-of-the-week effect on volatility includes French and Roll (1986), Berument and Kiymaz (2001) for the stock market, Harvey and Huang (1991), Ederington and Lee (1993) for interest rates and the foreign exchange futures market and Jones et al. (1998) for the Treasury bond market.
2

gas market during the winter may cause large price swings. This observation motivates my hypothesis that natural gas volatility is higher in winter. To my knowledge, this hypothesis has not been tested heretofore. While it is expected that natural gas volatility increases in the winter months, the answer is less obvious for crude oil volatility. On one hand, since part of U.S. oil consumption is for transportation, oil volatility supposedly increases in the summer when travel demand varies. On the other hand, the demand for heating oil, a product distilled from crude oil, may increase sharply in the winter3, and thus, crude oil volatility should increase in the winter months. Therefore, whether there is seasonality in crude oil volatility is subject to empirical evidence. 5. Do the Weekly Petroleum Status Report and the Weekly Natural Gas Storage Report releases cause increased crude oil and natural gas volatilities on announcement days? It has been documented that prices in other markets are generally more volatile when lots of new information is coming to the market. Ederington and Lee (1993, 1995) found that following the releases of scheduled macroeconomic news such as the employment report, the consumer price index (CPI) and the producer price index (PPI), interest rate and exchange rate volatilities are considerably higher than normal. Jones et al. (1998) found that the releases of employment and PPI data are responsible for an increase in volatility in the Treasury bond market on announcement days. Flannery and Protopapadakis (2002) found that 3 macroeconomic announcements significantly increase volatility in the stock market when they are released. I expect that consistent with the evidence from other markets, volatilities increase on days the Weekly Petroleum Status Report and the Weekly Natural Gas Storage Report4, which are reportedly among the most important news influencing the crude oil and natural gas markets, are released. To my knowledge no previous study has examined the possible impact of any scheduled announcement on oil volatility. The impact of the Storage Report announcement on natural gas volatility was first documented in Linn and Zhu (2004) but they use a Generalized Method of Moments (GMM) framework whereas I employ a GARCH (1,1) type model. 6. If there is evidence that crude oil and natural gas volatilities increase on scheduled announcement days, does the high volatility persist on days following the announcements? In other words, is public information in the reports immediately incorporated in oil and gas prices or does the announcement impact persist on subsequent days? If oil and gas market participants finish adjusting prices according to the new information within the announcement day, volatility should fall back to normal levels on the following day. On the other hand, GARCH type models consistently find evidence of volatility persistence, implying that if an announcement creates high volatility one day, that high volatility will tend to linger.

3 Heating oil can be used as a substitute for natural gas. Of the 107 million households in the United States, approximately 8.1 million use heating oil as their main heating fuel. Source: Energy Information Administration. 4 The Weekly Petroleum Status Report and the Weekly Natural Gas Storage Report are both compiled and issued by the U.S. Energy Information Administration.

The data in this study are the daily closing prices for the nearby crude oil and natural gas futures contracts traded on the New York Mercantile Exchange (NYMEX) from January 1, 1997 to July 31, 2007 totaling 2,645 daily observations. In order to explore the above questions (1-5), I estimate a model in which the conditional variance follows a regime-switching GJR type process which is similar to the model outlined in Jones et al. (1998): Rt = + 1Rt-1 + st1/2t, ht = + t-12+ht-1+ t-12It-1, st = {(1+ D,iDi,t)}(1+WWINt)(1+SSUMt)(1+ADAt), (1) (2) (3)

where Rt is the log percentage change in price of the nearby futures contract on day t, t is presumed to be an independent random variable which is normally distributed with conditional mean zero and conditional variance ht. It-1=1 if t-1 <0 and 0 otherwise. st is the transitory effects equation in which Di,t are the weekday dummies, WINt =1 in the winter months (from November through March) and 0 otherwise, SUMt =1 in the summer months (from June through August) and 0 otherwise. DAt is the announcement dummy =1 on days the reports are released. D,i, W, S and A measure day-of-the-week, seasonality and announcement effects on the conditional volatility This model, which separates conditional volatility into a persistent part (equation 2) and a transitory part (equation 3), allows conditional volatility to differ on each day of the week, in different times of the year and on announcement days. Furthermore, the st equation allows weekday and announcement effects impact that days volatility only. The above model is a methodological improvement over a number of studies on volatility in other markets5. In previous studies, the dummies are in the conditional variance equation (equation 2) and thus, weekday or announcement effect is forced to have persistent impact on conditional volatility on the following day6. In my study, the separation of conditional volatility into persistent and transitory parts allows me to implement a much cleaner study of the determinants of volatility. My estimation results from the specification (1-3) show that and are significantly positive, implying that volatility persistence is an attribute of the crude oil and natural gas markets. For the crude oil market, is significantly positive, indicating that a negative oil shock has more impact on predicted volatility than an equivalent positive shock. For the natural gas market, is significantly negative, indicating that a
See, for example, Hsieh (1989), Berument and Kiymaz (2001) and Ederington and Lee (2001) In previous studies, there is no st equation (equation 3) and weekday and/or announcement dummies are in the ht equation (equation 2): ht+1 = + t2 + ht + t2It+ MDWM,t+1 + WDWW,t+1 + RDWR,t+1 + FDWF,t+1, (4). In equation (4), the dummy for any day of the week impacts volatilities on all days of the week through the ht-1 term on the right hand side of the equation. Suppose, for instance, that day t is Monday. ht/DWM,t = M. Now consider the impact of the Monday dummy on the Tuesdays (day t+1) volatility. From equation (4), ht+1/DWM,t = M. Likewise, the Monday dummy impact on the Wednesdays volatility is ht+2/DWM,t = 2M. Thus, when weekday dummies are in the ht equation, as in equation (4), M does not measure how much higher volatility is on Monday than on the omitted day (Tuesday). Indeed, depending on the coefficient pattern, day X which has the highest X coefficient may not be the day with the highest volatility.
6 5

negative gas shock has less impact on the predicted volatility than a positive shock of the same magnitude. In both markets, the Friday-close-to-Monday-close volatility, which incorporates volatility over the weekend, is higher than any normal weekday volatility. The standard deviations of the Friday-close-to-Monday-close oil and gas returns increase by 20.08% and 39.39%, respectively. There is significant evidence that crude oil and natural gas volatilities increase in the winter months, confirming the seasonality effects in both markets. Crude oil and natural gas volatility significantly increases on days the Petroleum Report (for crude oil market) and the Storage Report (for natural gas market) releases are announced. In a separate test I find no significant evidence that the increase in oil and gas volatility due to announcement effects persists on the following days. III. Essay 2: Implied Volatility in the Crude oil and Natural gas Markets

In the second part of my dissertation, I examine the attributes of crude oil and natural gas implied volatilities calculated from daily closing prices of futures and futures options traded on the NYME from September 01, 1999 through June 30, 2006. The samples are broken into four time-to-maturity groups according to option expiry: near-, second-, third- and fourth- month. Each of the maturity group is then divided into ten moneyness subsamples corresponding to the amount the options are in- or out- of the money. In this essay, I attempt to answer the following questions: 1. Is there a term structure in crude oil and natural gas implied volatility? It has been

documented that in the stock index option market, implied volatilities are lower for options at shorter maturities. 7 To my knowledge, no previous study explored the term structure of implied volatility in any futures option market, including energy market. Contrary to the evidence in the stock index option market, I find that implied volatilities are higher on near term oil and gas options than on longer term ones. Thus, average implied volatility on nearby options is higher than average implied volatility on second-month options which is, in turn, higher than that on third-month options, etc. The term structure of implied volatilities on crude oil and natural gas options may possibly be explained by the pattern of actual volatilities in oil and gas futures markets. Summary statistics of returns in oil and gas futures markets indicate that returns on nearby contracts are more volatile than those on second-month which are, in turn, more volatile than those on third-month, etc. This may be due to the fact that news such as OPEC policy changes, refinery outages (for crude oil) or storage shortage, unfavorable weather forecasts (for natural gas) seem to have more impact on near term contracts than on longer term ones, causing larger price changes for the former. As implied volatility supposedly represents the markets expectation of actual volatility, the term structure of implied volatility should be consistent with the pattern of the actual volatility and thus, implied volatilities on near term oil and gas options should be higher than those on longer term ones.

Park and Sears (1985), Becker and Tucker (1991), Dumas, Fleming and Whaley (1998)

2.

Is there a smile in crude oil and natural gas implied volatilities? Since implied volatilities

calculated from different strike options with the same expiration date represent the markets expectation of volatility over the same period, there should be no significant difference in those implied volatilities. However, contrary to this theory, previous studies document sizable and persistent cross-sectional differences in implied volatilities in other markets. Implied volatilities on stock and stock index options, for example, form a smile pattern8 prior to the October 1987 market crash and a smirk pattern9 after the crash10. Although the literature is replete with studies on the implied volatility smile in various financial markets, none of the previous studies, to the best of my knowledge, explores the possible smile pattern in any futures options, including crude oil and natural gas options. My study aims to fill this gap in our understanding. I find evidence of a skew pattern in crude oil and natural gas implied volatilities where implied volatilities are generally lowest when options strikes are low and increase monotonically with strikes. Interestingly this skew pattern is opposite to the sneer, or smirk pattern in the post-1987 stock and stock index option markets where implied volatilities monotonically decrease with strikes. One possible explanation for the skew pattern in crude oil and natural gas options is the hedging pressure in the oil and gas markets similarly to the argument in Bollen and Whaley (2004) for stock indices. Given that crude oil and natural gas are two of the most essential energy sources, it is extremely difficult for consumers to quickly reduce their oil and gas consumption, even when a sharp increase in price occurs. Consequently, there may be more oil and gas users hedging against a price increase than there are sellers hedging against a price decrease, pushing upward prices and implied volatilities on options with high strikes. If much trading in the oil and gas options is due to hedging, I expect to find that call volume would be higher at the high strike options and put volume would be higher at the low strikes, especially the former. Also, if the hedging pressure hypothesis holds, I expect implied volatilities at the high strikes to be higher on calls than on puts (although they should be close according to put-call parity). 3. How well do crude oil and natural gas implied volatilities predict future volatility? In other

words, I examine whether implied volatility on crude oil and natural gas options is an unbiased and efficient forecast of actual volatility. The volatility implied in an options price is widely regarded as the markets forecast of future returns volatility over the remaining life of the option. If option markets are efficient, implied volatility should be an efficient forecast of future volatility, i.e., implied volatility should subsume all other information in explaining future volatility. The literature is replete with studies on whether implied volatility predicts future volatility

In a volatility smile, options that are deep in-the-money or out-of-the-money have higher implied volatilities than at-the-money options. 9 In a volatility smirk, implied volatilities decrease monotonically as the exercise price increases. 10 See, for example, Canina and Figlewski (1993), Rubinstein (1994), Dumas, Fleming and Whaley (1998), Das and Sundaram (1999), Ederington and Guan (2005)

and whether it does so efficiently in various markets, including the stock and stock index options market11, foreign exchange options market12, futures options markets13, Eurodollar options market14, etc. Day and Lewis (1993), Martens and Zein (2004) and Szakmary et al. (2003) examine the information content of crude oil and natural gas implied volatilities but their data are limited to implied volatilities calculated from nearby at-the-money options. In contrast, the large data set in this study enables me to further explore whether implied volatility calculated from any specific maturity and moneyness group is the best forecast of future volatility15. I expect implied volatilities on near term options to be the best forecast of future volatility. Furthermore, I expect that implied volatilities of high strike options whose prices are supposedly impacted heavily by hedging pressures should be less representative of the markets volatility expectation than implied volatilities calculated from options with lower strikes whose prices should be less subject to hedging pressures. I estimate the following specification on each maturity and moneyness subsample:

t ( ) = + 1 ISDi , j , t + ui , j ,t ,
t ( ) = ' + 1' ISDi , j , t + 2' HISi , j , t + ui , j , t ,

(1) and (2)

where ISDi , j ,t is the implied volatility (measured as the implied standard deviation) computed on day t from the option in maturity group i and moneyness group j , and ui , j , t represents the regression error.

t ( ) is

the realized volatility of log returns over the period between t and t + , the options expiration date, annualized by multiplying the standard deviation calculated per day by

252. Log return is defined as:

F Rt = Ln t where Ft and Ft-1 are the underlying futures prices on day t and day t-1, respectively. Ft 1
I then employ Hansen (1982) procedure to correct for serial correlation caused by overlapping observations. Results from specification (1) and (2) show that implied volatilities calculated from near-the-money and slightly out-of-the-money call options in the nearby and second-month groups are unbiased and efficient predictors of future volatility in the crude oil and natural gas markets.

See, for example, Day and Lewis (1992), Lamoureux and Lastrapes (1993). Canina and Figlewski (1993), Christensen and Prabhala (1998), etc. 12 Jorion (1995) 13 See, Day and Lewis (1993), Martens and Zein (2004), Szakmary, Ors and Kim (2003) 14 Amin and Ng (1997) 15 This question is inspired by the findings in Ederington and Guan (2005) that implied volatilities calculated from moderately high strike options are both unbiased and efficient whereas those on at-the-money options are biased and less efficient.

11

4.

Is there a seasonality pattern in crude oil and natural gas implied volatilities? As analyzed in

the first essay, a winter effect is an attribute of the natural gas market due to the dependence of natural gas prices on weather conditions, and to a lesser extent, there is also evidence of higher crude oil volatility in winter. In this essay, I hypothesize that implied volatilities are higher on oil and gas options expiring in the winter months than on options expiring in other months. This hypothesis is likely due to two reasons. First, as the market participants forecast actual volatilities to increase in winter, this forecast is incorporated in options prices and thus, implied volatilities are higher on options expiring in the winter months. Second, as oil and gas volatilities tend to increase in winter, there may be more oil and gas users buying options to hedge against price increases, pushing upward prices and implied volatilities on options expiring in the winter months. To my knowledge, none of the previous studies examined seasonality pattern for any option market. I find evidence of a seasonality pattern in crude oil and natural gas option markets in that average implied volatilities on options expiring in the winter months are higher than those on options expiring in other months. In the natural gas option market, the average implied volatility (as measured by implied standard deviation) calculated from options expiring in the winter peak is approximately 60% higher than those expiring in the summer trough whereas in the crude oil option market, the peak is 22% higher than the trough. This month-of-the-year pattern is consistent across all maturity groups in both markets. 5. Is there a day-of-the-week pattern in crude oil and natural gas implied volatilities? A day-of-

the-week effect on implied volatility has been documented for several markets16. However, no previous study explored weekday pattern in any futures option market, including crude oil and natural gas. I find that implied volatilities in crude oil and natural gas option markets significantly increase on Friday and decrease on Monday. As the forecast of Monday actual volatility is included in Friday implied volatility (which represents the markets expectation of future volatility from the following Monday to the option expiration) and dropped from Monday implied volatility (which represents the markets expectation of future volatility from Tuesday to the option expiration), evidence of a higher Friday implied volatility and a lower Monday implied volatility indicates that Monday is a high volatile day in both markets. 6. Do the crude oil and natural gas implied volatilities tend to fall following important scheduled

announcements? In the first essay, I find that crude oil and natural gas actual volatilities increase on days the Weekly Petroleum Status Report and the Weekly Natural Gas Storage Report are announced. Consistent with this finding, I expect that crude oil and natural gas implied volatilities tend to fall following these announcements. Since market participants know that important scheduled releases tend to impact the underlying prices, uncertainty, as measured by implied volatility, should be high prior to the release. Following the announcement release, the actual volatility should increase but the implied volatility should fall as this source of uncertainty is resolved. (Ederington and Lee (1996)).
16

See, for example, Becker and Tucker (1991), Harvey and Whaley (1992) and Ederington and Lee (1996)

7.

Do positive and negative shocks in crude oil and natural gas returns have different impact on

the subsequent implied volatility? In the first essay, I find that a negative shock in the crude oil and natural gas futures markets has a different impact on predicted volatility than a positive shock of the same magnitude. Consequently, in this essay, I investigate whether a negative shock in these markets has more or less impact on the implied volatility than an equal positive shock. The main difference between this part and the previous part is that the predicted volatility in the futures markets is forecast from a GJR type model17 whereas in this part, implied volatility is the forecast of future volatility. This is the first time the impact of underlying returns shocks on the subsequent implied volatility is explored in the literature. To test the impact of positive and negative returns shocks on implied volatility, I use the following specification: ISDt = 0 + 1 ISDt 1 + 2 t 1 + 3t 1 t 1 + t (3) where ISDt is the implied standard deviation on day t, t = t 260 and t is the residual from the following equation: Rt = + Rt 1 + t (4) ;

Rt = ln( Ft / F

t 1

) where Ft is the futures price on day t, t =1 if t <0 and 0 otherwise.

To the best of my knowledge, my second essay will be the first full study of the attributes of implied volatility such as the term structure and the smile in implied volatility, information content, seasonality, dayof-the-week and announcement effects and the impact of previous returns shocks on implied volatility. Reference Adrian R., G. Pagan and W. Schwert, 1990, Alternative Models for Conditional Stock Volatility, Journal of Econometrics 45, 267-290. Amin K. and V. Ng , 1997, Inferring Future Volatility from the Information in Implied Volatility in Eurodollar Options: A New Approach, Review of Financial Studies 10, 333367. Andersen T.G., and T. Bollerslev, 1998, Deutsche Mark-Dollar Volatility: Intraday Activity Patterns, Macroeconomic Announcements, and Longer Run Dependencies, Journal of Finance 53, 219-265. Andersen T.G., T. Bollerslev, F.X. Diebold and H. Ebens, 2001, The Distribution of Realized Stock Return Volatility, Journal of Financial Economics 61, 43-76. Bekaert G., and G.J Wu, 2000, Asymmetric Volatility and Risk in Equity Markets, Review of Financial Studies 13, 1-42. Berument H., and H. Kiymaz, 2001, The Day of the Week Effect on Stock Market Volatility, Journal of Economics and Finance 25, 181 - 193. Bollen N.P. and R.E. Whaley, 2004, Does Net Buying Pressure affect the Shape of Implied Volatility Functions?, Journal of Finance 59, 711-753. Bollerslev T., 1986, Generalized Autoregressive Conditional Heteroskedasticity, Journal of econometrics 31, 307-327.

17

The GJR model is named after Glosten, Jagannathan and Runkle (1993)

10

Bollerslev T., RY. Chou and K.F. Kroner, 1992, ARCH Modeling in Finance: A Review of the Theory and Empirical Evidence, Journal of Econometrics 52, 5-59. Brunner A.D., and D.P. Simon, 1996, Excess Returns and Risk at the long End of The Treasury Market: an EGARCH-M Approach, The Journal of Financial Research 14, 443-457. Campbell J.Y., and L. Hentschel, 1992, No News is Good News: An Asymmetric Model of Changing Volatility in Stock Returns, Journal of Financial Economics 31, 279-411. Canina L. and S. Figlewski, 1993, The Informational Content of Implied Volatility, Review of Financial Studies 6, 659-681. Christensen B.J. and N.R. Prabhala, 1998, The Relation between Implied and Realized Volatility, Journal of Financial Economics 50, 125-150. Das S.R. and R.K.Sundaram, 1999, Of Smiles and Smirks: A Term Structure Perspective, The Journal of Financial and Quantitative Analysis 34, 211-239. Day T.E. and C.M. Lewis, 1992, Stock Market Volatility and the Information Content of Stock Index Options, Journal of Econometrics 52, 267287. Day T.E., and C.M. Lewis, 1993, Forecasting Futures Market Volatility, Journal of Derivatives 1, 33-50. De Goeij P., and W. Marquering, 2006, Macroeconomic Announcements and Asymmetric Volatility in Bond Returns, Journal of Banking and Finance 30, 2659-2680. Dumas B., Fleming J. and R.E.Whaley, 1998, Implied Volatility Functions: Empirical Tests, The Journal of Finance 53, 20592106. Ederington L., and J.H. Lee, 1993, How Markets Process Information: New Releases and Volatility, Journal of Finance 48, 1161-1191. Ederington L., and J.H. Lee, 1995, The Short-run Dynamics of the Price Adjustment to New Information, Journal of Financial and Quantitative Analysis 30, 117-134. Ederington L. H. and J. H. Lee, 1996, The Creation and Resolution of Market Uncertainty: The Impact of Information Releases on Implied Volatility, Journal of Financial and Quantitative Analysis 31, 513 539. Ederington L., and J.H. Lee, 2001, Volatility in Interest-rate and Foreign-exchange Markets: ARCH, Announcement, and Seasonality Effects, Journal of Futures Markets 21, 17-552. Ederington L.H. and W.Guan, 2005, The Information Frown in Option Prices, Journal of Banking & Finance 29, 1429-1457. Engle R.F., 1982, Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom inflation, Econometrica 50, 987-1982. Engle R.F., and V.K. Ng, 1993, Measuring and Testing the Impact of News on Volatility, Journal of Finance 48, 1749-1778. Engle R.F., 2001, GARCH 101: The Use of ARCH/GARCH Models in Applied Econometrics, The Journal of Economic Perspectives 15, 157-168. Flannery M.J., and A.A. Protopapadakis, 2002, Macroeconomic Factors Do Influence Aggregate Stock Returns, Review of Financial Studies 15, 751-782. 11

French K.R., and R. Roll, 1986, Stock Returns Variances: The Arrival of Information and the Reaction of Traders, Journal of Financial Economics 17, 5-26. French K.R., G.W. Schwert and R.F. Stambaugh, 1987, Expected Stock Returns and Volatility, Journal of Financial Economics 19, 3-29. Glosten L.R., R. Jagannathan and D.E. Runkle, 1993, On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks, Journal of Finance 48, 1779-1801. Harvey C.R., and R.D. Huang, 1991, Volatility in the Foreign-currency Futures Market, Review of Financial Studies 4, 543-569. Harvey C.R. and R.E. Whaley, 1992, Market Volatility Prediction and the Efficiency of the S&P 100 Index Option Market, Journal of Financial Economics 31, 4373. Hsieh D.A., 1989, Modeling Heteroskedasticity in Daily Foreign Exchange Rates, Journal of Business and Economic Statistics 7, 307-317. Jones C.M., O. Lamont and R.L. Lumsdaine, 1998, Macroeconomic News and Bond Market Volatility, Journal of Financial Economics 47, 315-337. Jordan S.D., and B.D. Jordan, 1991, Seasonality in Daily Bond Returns, The Journal of Financial and Quantitative Analysis. 26, 269-285. Jorion P., 1995, Predicting Volatility in the Foreign Exchange Market, The Journal of Finance 50, 507-528. Keim D.B., 1983, Size Related Anomalies and Stock Return Seasonality: Further Empirical Evidence, Journal of Financial Economics 12, 13-32. Kim S., 1999, Do Macro-economic News Announcements Affect the Volatility of Foreign Exchange Rates? Some Evidence from Australia, Applied Economics 31, 1511-1521. Kuper G.H. and D.P. van Soest, 2006, Does Oil Price Uncertainty affect Economic Activity?, Energy Journal 27, 55-78. Kwek K.T., and K.N Koay, 2006, Exchange Rate Volatility and Volatility Asymmetries: an Application to Finding a Natural Dollar Currency, Applied Economics 38, 307-323. Lakonishok J., and S. Smidt, 1984, Volume and Turn-of-the-year Behavior, Journal of Financial Economics 13, 435-455. Lamoureux C.G., and W.D. Lastrapes, 1993, Forecasting Stock Return Variance: Toward an Understanding of Stochastic Implied Volatilities, Review of Financial Studies 6, 293-326. Linn S.C., and Z. Zhu, 2004, Natural Gas Prices and the Gas Storage Report: Public News and Volatility in Energy Futures Markets, The Journal of Futures Markets 24, 283- 313. Low B.S., and S. Zhang, 2005, The Volatility Risk Premium Embedded in Currency Options, Journal of Financial and Quantitative Analysis 40, 803-832. Martens, M., and J. Zein, 2004, Predicting financial volatility: High-frequency time-series forecasts vis--vis implied volatility, The Journal of Futures Markets 2004, 1005-1029. McKenzie M.D., 2002, The Economics of Exchange Rate Volatility Asymmetry, International Journal of Finance and Economics 7, 247260. 12

Nelson D.B., 1991, Conditional Heteroskedasticity in Asset Returns: A New Approach, Econometrica 59, 347-370. Park H.Y and S.R. Sears, 1985, Estimating Stock Index Futures Volatility through the Prices of Their Options, The Journal of Futures Markets 5, 223-238 Petak K., 2004, Domestic Production and Demand Forecast, Presentation at the 2004 Winter Fuels Outlook Conference, Washington, D.C. Pindyck R.S, Volatility in Natural Gas and Oil Markets, 2004, The Journal of Energy and Development 30, 119. Poon S.H., and C.W.J. Granger, 2003, Forecasting Volatility in Financial Markets: A Review, Journal of Economic Literature 41, 478-539. Rubinstein M., 1994, Implied binomial trees, Journal of Finance 49, 771-818. Szakmary A.E., Ors E., Kim J. and W. Davidson, 2003, The Predictive Power of Implied Volatility: Evidence from 35 Futures Markets, Journal of Banking and Finance 27, 21512176 Simon D., 1997, Implied Volatility Asymmetries in Treasury Bond Futures Options, Journal of Futures Markets 17, 873-885. Susmel R., and A. Thompson, 1997, Volatility Storage and Convenience: Evidence from Natural Gas Markets, Journal of Futures Markets 17, 17-43. Tucker A. and K.G. Becker, 1991, Implied Index Volatilities and Intraweek Patterns in the U.S. Equity Market, Advances in Futures and Options Research 5, 297-308 U.S. Energy Information Administration, 2005, An Assessment of Prices of Natural Gas Futures Contracts as a Predictor of Realized Spot Prices at the Henry Hub. Veronesi P., 1999, Stock Market Overreaction to Bad News in Good Times: A Rational Expectations Equilibrium Model, Review of Financial Studies 12, 975-1007. Wilson B., R. Aggarwal and C. Inclan, 1996, Detecting Volatility Changes Across the Oil Sector, Journal of Futures Markets 16, 313-330. Wu G.J., 2001, The Determinants of Asymmetric Volatility, Review of Financial Studies 14, 837-859. Yang C. W., M. Hwang, and B. N. Huang, 2002, An analysis of factors affecting price volatility of the U. S. oil market, Energy Economics 24, 107-119.

13