Simulation for Data Science with R
()
About this ebook
- Learn five different simulation techniques (Monte Carlo, Discrete Event Simulation, System Dynamics, Agent-Based Modeling, and Resampling) in-depth using real-world case studies
- A unique book that teaches you the essential and fundamental concepts in statistical modeling and simulation
This book is for users who are familiar with computational methods and R. If you want to learn about the advanced features of R, with the computer-intense Monte Carlo methods and tools for statistical simulation, then this book is for you.
Related to Simulation for Data Science with R
Related ebooks
Data Mining Applications with R Rating: 4 out of 5 stars4/5Big Data Analytics with R Rating: 0 out of 5 stars0 ratingsBayesian Analysis with Python Rating: 5 out of 5 stars5/5Python Data Science Essentials Rating: 0 out of 5 stars0 ratingsLearning Predictive Analytics with R Rating: 0 out of 5 stars0 ratingsHandbook of Statistical Analysis and Data Mining Applications Rating: 4 out of 5 stars4/5Mastering Predictive Analytics with R Rating: 4 out of 5 stars4/5Learning Bayesian Models with R Rating: 5 out of 5 stars5/5Mastering Data Analysis with R Rating: 5 out of 5 stars5/5Mastering Text Mining with R Rating: 0 out of 5 stars0 ratingsLearning Probabilistic Graphical Models in R Rating: 0 out of 5 stars0 ratingsR for Data Science Rating: 5 out of 5 stars5/5R: Data Analysis and Visualization Rating: 5 out of 5 stars5/5Regression Analysis with Python Rating: 0 out of 5 stars0 ratingsPractical Data Analysis Rating: 4 out of 5 stars4/5Principles of Data Science Rating: 4 out of 5 stars4/5Regression Analysis: An Intuitive Guide for Using and Interpreting Linear Models Rating: 5 out of 5 stars5/5Machine Learning with R - Second Edition Rating: 5 out of 5 stars5/5Mastering Scientific Computing with R Rating: 3 out of 5 stars3/5Building a Recommendation System with R Rating: 0 out of 5 stars0 ratingsPython Data Analysis - Second Edition Rating: 0 out of 5 stars0 ratingsMachine Learning with R Rating: 4 out of 5 stars4/5R High Performance Programming Rating: 4 out of 5 stars4/5Learning Data Mining with Python Rating: 0 out of 5 stars0 ratingsMastering Python for Data Science Rating: 3 out of 5 stars3/5Python Data Science Essentials - Second Edition Rating: 4 out of 5 stars4/5Mastering Machine Learning with R Rating: 0 out of 5 stars0 ratingsIntroduction to R for Business Intelligence Rating: 0 out of 5 stars0 ratingsPython Data Analysis Rating: 4 out of 5 stars4/5Learning Predictive Analytics with Python Rating: 0 out of 5 stars0 ratings
Data Visualization For You
Visual Analytics with Tableau Rating: 0 out of 5 stars0 ratingsData Analytics for Beginners: Introduction to Data Analytics Rating: 4 out of 5 stars4/5Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals Rating: 4 out of 5 stars4/5Teach Yourself VISUALLY Power BI Rating: 0 out of 5 stars0 ratingsHow to Become a Data Analyst: My Low-Cost, No Code Roadmap for Breaking into Tech Rating: 0 out of 5 stars0 ratingsHands-On Data Analysis with Pandas: Efficiently perform data collection, wrangling, analysis, and visualization using Python Rating: 0 out of 5 stars0 ratingsHow to Lie with Maps Rating: 4 out of 5 stars4/5Learning Tableau 10 - Second Edition Rating: 4 out of 5 stars4/5Learning pandas - Second Edition Rating: 4 out of 5 stars4/5Data Visualization: A Practical Introduction Rating: 5 out of 5 stars5/5Learn D3.js: Create interactive data-driven visualizations for the web with the D3.js library Rating: 0 out of 5 stars0 ratingsR for Data Science Rating: 5 out of 5 stars5/5The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios Rating: 4 out of 5 stars4/5DAX Patterns: Second Edition Rating: 5 out of 5 stars5/5Data Analysis with Stata Rating: 5 out of 5 stars5/5Excel for Beginners 2023: A Step-by-Step and Comprehensive Guide to Master the Basics of Excel, with Formulas, Functions, & Charts Rating: 0 out of 5 stars0 ratingsD3.js in Action: Data visualization with JavaScript Rating: 0 out of 5 stars0 ratingsClojure Data Analysis Cookbook - Second Edition Rating: 0 out of 5 stars0 ratingsMastering Text Mining with R Rating: 0 out of 5 stars0 ratings#MakeoverMonday: Improving How We Visualize and Analyze Data, One Chart at a Time Rating: 0 out of 5 stars0 ratingsVisualizing Graph Data Rating: 0 out of 5 stars0 ratingsHow to be Clear and Compelling with Data: Principles, Practice and Getting Beyond the Basics Rating: 0 out of 5 stars0 ratingsMastering Data Analysis with Python: A Comprehensive Guide to NumPy, Pandas, and Matplotlib Rating: 0 out of 5 stars0 ratingsGoogle Analytics 4 Migration Quick Guide 2022: Universal Analytics disappears in July 2023 - are you ready? Rating: 0 out of 5 stars0 ratingsArchitecting Big Data & Analytics Solutions - Integrated with IoT & Cloud Rating: 5 out of 5 stars5/5
Reviews for Simulation for Data Science with R
0 ratings0 reviews
Book preview
Simulation for Data Science with R - Matthias Templ
Table of Contents
Simulation for Data Science with R
Credits
About the Author
About the Reviewer
www.PacktPub.com
eBooks, discount offers, and more
Why subscribe?
Preface
What this book covers
What you need for this book
Who this book is for
Conventions
Reader feedback
Customer support
Downloading the example code
Downloading the color images of this book
Errata
Piracy
Questions
1. Introduction
What is simulation and where is it applied?
Why use simulation?
Simulation and big data
Choosing the right simulation technique
Summary
References
2. R and High-Performance Computing
The R statistical environment
Basics in R
Some very basic stuff about R
Installation and updates
Help
The R workspace and the working directory
Data types
Vectors in R
Factors in R
list
data.frame
array
Missing values
Generic functions, methods, and classes
Data manipulation in R
Apply and friends with basic R
Basic data manipulation with the dplyr package
dplyr – creating a local data frame
dplyr – selecting lines
dplyr – order
dplyr – selecting columns
dplyr – uniqueness
dplyr – creating variables
dplyr – grouping and aggregates
dplyr – window functions
Data manipulation with the data.table package
data.table – variable construction
data.table – indexing or subsetting
data.table – keys
data.table – fast subsetting
data.table – calculations in groups
High performance computing
Profiling to detect computationally slow functions in code
Further benchmarking
Parallel computing
Interfaces to C++
Visualizing information
The graphics system in R
The graphics package
Warm-up example – a high-level plot
Control of graphics parameters
The ggplot2 package
References
3. The Discrepancy between Pencil-Driven Theory and Data-Driven Computational Solutions
Machine numbers and rounding problems
Example – the 64-bit representation of numbers
Convergence in the deterministic case
Example – convergence
Condition of problems
Summary
References
4. Simulation of Random Numbers
Real random numbers
Simulating pseudo random numbers
Congruential generators
Linear and multiplicative congruential generators
Lagged Fibonacci generators
More generators
Simulation of non-uniform distributed random variables
The inversion method
The alias method
Estimation of counts in tables with log-linear models
Rejection sampling
Simulating values from a normal distribution
Simulating random numbers from a Beta distribution
Truncated distributions
Metropolis - Hastings algorithm
A few words on Markov chains
The Metropolis sampler
The Gibbs sampler
The two-phase Gibbs sampler
The multiphase Gibbs sampler
Application in linear regression
The diagnosis of MCMC samples
Tests for random numbers
The evaluation of random numbers – an example of a test
Summary
References
5. Monte Carlo Methods for Optimization Problems
Numerical optimization
Gradient ascent/descent
Newton-Raphson methods
Further general-purpose optimization methods
Dealing with stochastic optimization
Simplified procedures (Star Trek, Spaceballs, and Spaceballs princess)
Metropolis-Hastings revisited
Gradient-based stochastic optimization
Summary
References
6. Probability Theory Shown by Simulation
Some basics on probability theory
Probability distributions
Discrete probability distributions
Continuous probability distributions
Winning the lottery
The weak law on large numbers
Emperor penguins and your boss
Limits and convergence of random variables
Convergence of the sample mean – weak law of large numbers
Showing the weak law of large numbers by simulation
The central limit theorem
Properties of estimators
Properties of estimators
Confidence intervals
A note on robust estimators
Summary
References
7. Resampling Methods
The bootstrap
A motivating example with odds ratios
Why the bootstrap works
A closer look at the bootstrap
The plug-in principle
Estimation of standard errors with bootstrapping
An example of a complex estimation using the bootstrap
The parametric bootstrap
Estimating bias with bootstrap
Confidence intervals by bootstrap
The jackknife
Disadvantages of the jackknife
The delete-d jackknife
Jackknife after bootstrap
Cross-validation
The classical linear regression model
The basic concept of cross validation
Classical cross validation – 70/30 method
Leave-one-out cross validation
k-fold cross validation
Summary
References
8. Applications of Resampling Methods and Monte Carlo Tests
The bootstrap in regression analysis
Motivation to use the bootstrap
The most popular but often worst method
Bootstrapping by draws from residuals
Proper variance estimation with missing values
Bootstrapping in time series
Bootstrapping in the case of complex sampling designs
Monte Carlo tests
A motivating example
The permutation test as a special kind of MC test
A Monte Carlo test for multiple groups
Hypothesis testing using a bootstrap
A test for multivariate normality
Size of the test
Power comparisons
Summary
References
9. The EM Algorithm
The basic EM algorithm
Some prerequisites
Formal definition of the EM algorithm
Introductory example for the EM algorithm
The EM algorithm by example of k-means clustering
The EM algorithm for the imputation of missing values
Summary
References
10. Simulation with Complex Data
Different kinds of simulation and software
Simulating data using complex models
A model-based simple example
A model-based example with mixtures
Model-based approach to simulate data
An example of simulating high-dimensional data
Simulating finite populations with cluster or hierarchical structures
Model-based simulation studies
Latent model example continued
A simple example of model-based simulation
A model-based simulation study
Design-based simulation
An example with complex survey data
Simulation of the synthetic population
Estimators of interest
Defining the sampling design
Using stratified sampling
Adding contamination
Performing simulations separately on different domains
Inserting missing values
Summary
References
11. System Dynamics and Agent-Based Models
Agent-based models
Dynamics in love and hate
Dynamic systems in ecological modeling
Summary
References
Index
Simulation for Data Science with R
Simulation for Data Science with R
Copyright © 2016 Packt Publishing
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews.
Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing, and its dealers and distributors will be held liable for any damages caused or alleged to be caused directly or indirectly by this book.
Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information.
First published: June 2016
Production reference: 1240616
Published by Packt Publishing Ltd.
Livery Place
35 Livery Street
Birmingham B3 2PB, UK.
ISBN 978-1-78588-116-9
www.packtpub.com
Credits
Author
Matthias Templ
Reviewer
Gerlinde Dinges
Commissioning Editor
Akram Hussain
Acquisition Editor
Vinay Argekar
Content Development Editor
Aishwarya Pandere
Technical Editor
Tanmayee Patil
Copy Editors
Safis Editing
Angad Singh
Project Coordinator
Nidhi Joshi
Proofreader
Safis Editing
Indexer
Monica Ajmera Mehta
Graphics
Jason Monteiro
Production Coordinator
Nilesh Mohite
Cover Work
Nilesh Mohite
About the Author
Matthias Templ is associated professor at the Institute of Statistics and Mathematical Methods in Economics, Vienna University of Technology (Austria). He is additionally employed as a scientist at the methods unit at Statistics Austria, and together with two colleagues, he owns the company called data-analysis OG. His main research interests are in the areas of imputation, statistical disclosure control, visualization, compositional data analysis, computational statistics, robustness teaching in statistics, and multivariate methods. In the last few years, Matthias has published more than 45 papers in well-known indexed scientific journals. He is the author and maintainer of several R packages for official statistics, such as the R package sdcMicro for statistical disclosure control, the VIM package for visualization and imputation of missing values, the simPop package for synthetic population simulation, and the robCompositions package for robust analysis of compositional data. In addition, he is the editor of the Austrian Journal of Statistics that is free of charge and open-access. The probability is high to find him at the top of a mountain in his leisure time.
I'd like to thank my wife, Barbara, for her mental support for the work on this book. In addition, my thanks goes to all the people who provide open source code to the R project.
About the Reviewer
Gerlinde Dinges works as a methodological researcher at the quality management and methods department of Statistics Austria. She studied at the University of Linz and earned a degree in statistics from the University of Vienna. Since 2002, she has been working extensively in the field of model-based estimation for business statistics and published several papers in this field. Gerlinde has broad experience in working with large (administrative) data and has participated in several European research projects. Her main research interest is in the area of explorative data analysis, business statistics, imputation, visualization, and teaching statistics. Together with her colleagues, she developed a computer-assisted teaching system to enhance and support classroom teaching in an interactive way (http://www.statistik.at/tgui).
www.PacktPub.com
eBooks, discount offers, and more
Did you know that Packt offers eBook versions of every book published, with PDF and ePub files available? You can upgrade to the eBook version at www.PacktPub.com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touch with us at
At www.PacktPub.com, you can also read a collection of free technical articles, sign up for a range of free newsletters and receive exclusive discounts and offers on Packt books and eBooks.
https://www2.packtpub.com/books/subscription/packtlib
Do you need instant solutions to your IT questions? PacktLib is Packt's online digital book library. Here, you can search, access, and read Packt's entire library of books.
Why subscribe?
Fully searchable across every book published by Packt
Copy and paste, print, and bookmark content
On demand and accessible via a web browser
Preface
"Everybody seems to think I'm lazy
I don't mind, I think they're crazy
Running everywhere at such a speed
Till they find there's no need (There's no need)"
The Beatles in their song I'm only sleeping
The Monte Carlo way and simulation approach are ways to stay lazy and efficient at the same time. Lazy
, since a simulation approach is generally much easier to carry out as compared to an analytical approach—there is mostly no need for analytical approaches, and one might be crazy to neglect the whole world of statistical simulation. Efficient
, since it costs minimal efforts to get reliable results, and often simulation is the only approach to get results. The simulation approach in data science and statistics is generally a more intuitive approach compared to analytical solutions. It is not hidden behind a wall of mathematics, and using a simulation approach is often the only way to solve complex problems.
Statistical simulation has thus become an essential area in data science and statistics. It can be seen as a data-driven approach to many practical problems in data science and statistics.
In this book, theory is also explained with illustrative examples using the software environment R, for which advanced data processing features are shown in the book.
This book will thus provide a computational and methodological framework for statistical simulation to users with a computational statistics and/or data science background.
More precisely, the aim of this book is to lay into the hands of the readers a book that explains methods, give advice on the usage of the methods, and provide computational tools to solve common problems in statistical simulation and computer-intense methods.
The core issues are on simulating distributions and datasets, Monte Carlo methods for inference statistics, microsimulation and dynamical systems, and presenting solutions using computer-intense approaches. You will see applications in R not only to better understand the methods but also to gain experience when working on real-world data and real-world problems.
The author of the book has tried to make humorous and amusing examples in certain chapters in order to increase interest, staying catchy and memorable. Next to serious text on methods, curious examples on individual mortality and fertility rates of the author of the book are also present as is the system dynamics from the love/hate story of Prince Henry and Chelsy Davy, the Australian guy in the Austrian mountain trying to reach the highest mountain through an optimization problem, or the weak law of winning the lottery are presented as well.
What this book covers
Chapter 1, Introduction, discusses the general aim of simulation experiments in data science and statistics, why and where simulation is used, and the special case of dealing with big data.
Chapter 2, R and High-Performance Computing, consists of comprehensive text on advanced computing, data manipulation, and visualization with R.
Chapter 3, The Discrepancy between Pencil-Driven Theory and Data-Driven Computational Solutions, reports problems on numerical precision, rounding, and convergence in a deterministic setting.
Chapter 4, Simulation of Random Numbers, starts with the simulation of uniform random numbers and transformation methods to obtain other kinds of distributions. It includes a discussion of various types of Markov chain Monte Carlo (MCMC) methods.
Chapter 5, Monte Carlo Methods for Optimization Problems, introduces deterministic and stochastic optimization methods.
Chapter 6, Probability Theory Shown by Simulation, has a strong focus on basic theorems in statistics; for example, the concept of the weak law of large numbers and the central limit theorem are shown by simulation.
Chapter 7, Resampling Methods, is a comprehensive view on the bootstrap, the jackknife and cross-validation.
Chapter 8, Applications of Resampling Methods and Monte Carlo Tests, shows applications in various fields such as regression, imputation, and time series analysis. In addition, Monte Carlo tests and their variants such as permutation tests and bootstrap tests are presented.
Chapter 9, The EM Algorithm, introduces the expectation maximum method to iteratively obtain an optima. Applications in clustering and imputation of missing values are given.
Chapter 10, Simulation with Complex Data, shows how to simulate synthetic data as well as population data that can be used for the comparison of methods in general or also serve as input for agent-based microsimulation models.
Chapter 11, System Dynamics and Agent-Based Models, discusses agent-based microsimulation models and shows basic models in system dynamics to study complex dynamical systems.
What you need for this book
This book heavily depends on the software environment R, version 3.2 or newer (https://cran.r-project.org/). In most chapters, independent and standalone code is written to show methods and execute examples, and no additional packages of R are needed. For a few chapters, additional R packages such as deSolve, cvTools, laeken, VIM, and few others must be installed within R. The packages dplyr and ggplot2 are used throughout the book.
Optionally, the use of a script editor for R, such as RStudio (https://www.rstudio.com/) or Architect + Eclipse (https://www.openanalytics.eu/architect), is recommended.
Who this book is for
This book is for users who are familiar with computational methods and R. If you want to learn about the advanced features of R, along with computer-intense Monte Carlo methods and tools for statistical simulation, and if you prefer data-driven solutions, then this book is for you.
Conventions
In this book, you will find a number of text styles that distinguish between different kinds of information. Here are some examples of these styles and an explanation of their meaning.
R code words in text, filenames, file extensions, pathnames, and dummy URLs are shown as follows:
A block of code is set as follows:
[default]
love <- function(t, x, parms){
with(as.list(c(parms, x)), {
dPrince_Harry <- a * Chelsy_Davy
dChelsy_Davy <- -b * Prince_Harry
res <- c(dPrince_Harry, dChelsy_Davy)
list(res)
})
}
Any command-line input or output is written as follows:
dat <- matrix(c(104,11037,189,11034),2,2, byrow=TRUE) ## Loading required package: grid > confint(oddsratio(dat, log=FALSE)) ## 2.5 % 97.5 % ## 0.4324132 0.6998549
Note
Warnings or important notes appear in a box like this.
Tip
Tips and tricks appear like this.
Reader feedback
Feedback from our readers is always welcome. Let us know what you think about this book—what you liked or disliked. Reader feedback is important for us as it helps us develop titles that you will really get the most out of.
To send us general feedback, simply e-mail <feedback@packtpub.com>, and mention the book's title in the subject of your message.
If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, see our author guide at www.packtpub.com/authors.
Customer support
Now that you are the proud owner of a Packt book, we have a number of things to help you to get the most from your purchase.
Downloading the example code
You can download the example code files for this book from your account at http://www.packtpub.com. If you purchased this book elsewhere, you can visit http://www.packtpub.com/support and register to have the files e-mailed directly to you.
You can download the code files by following these steps:
Log in or register to our website using your e-mail address and password.
Hover the mouse pointer on the SUPPORT tab at the top.
Click on Code Downloads & Errata.
Enter the name of the book in the Search box.
Select the book for which you're looking to download the code files.
Choose from the drop-down menu where you purchased this book from.
Click on Code Download.
You can also download the code files by clicking on the Code Files button on the book's webpage at the Packt Publishing website. This page can be accessed by entering the book's name in the Search box. Please note that you need to be logged in to your Packt account.
Once the file is downloaded, please make sure that you unzip or extract the folder using the latest version of:
WinRAR / 7-Zip for Windows
Zipeg / iZip / UnRarX for Mac
7-Zip / PeaZip for Linux
The code bundle for the book is also hosted on GitHub at https://github.com/PacktPublishing/Simulation-for-Data-Science-with-R. We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!
Downloading the color images of this book
We also provide you with a PDF file that has color images of the screenshots/diagrams used in this book. The color images will help you better understand the changes in the output. You can download this file from https://www.packtpub.com/sites/default/files/downloads/SimulationforDataSciencewithR_ColorImages.pdf.
Errata
Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you find a mistake in one of our books—maybe a mistake in the text or the code—we would be grateful if you could report this to us. By doing so, you can save other readers from frustration and help us improve subsequent versions of this book. If you find any errata, please report them by visiting http://www.packtpub.com/submit-errata, selecting your book, clicking on the Errata Submission Form link, and entering the details of your errata. Once your errata are verified, your submission will be accepted and the errata will be uploaded to our website or added to any list of existing errata under the Errata section of that title.
To view the previously submitted errata, go to https://www.packtpub.com/books/content/support and enter the name of the book in the search field. The required information will appear under the Errata section.
Piracy
Piracy of copyrighted material on the Internet is an ongoing problem across all media. At Packt, we take the protection of our copyright and licenses very seriously. If you come across any illegal copies of our works in any form on the Internet, please provide us with the location address or website name immediately so that we can pursue a remedy.
Please contact us at <copyright@packtpub.com> with a link to the suspected pirated material.
We appreciate your help in protecting our authors and our ability to bring you valuable content.
Questions
If you have a problem with any aspect of this book, you can contact us at <questions@packtpub.com>, and we will do our best to address the problem.
Chapter 1. Introduction
In the previous century, the Vienna University of Technology in Vienna enrolled a bachelor study called data engineering and statistics. Basically the content was perfectly related to the nowadays commonly-used term data science. Data-oriented lectures in the area of computer science, such as storing and retrieving data, programming, and data security, were in the curriculum, together with applied lectures on statistics, such as multivariate statistics, biostatistics, financial statistics, statistical learning, and official statistics. We had too few students and after a few years the course was canceled. 16 years later, the picture completely changed. New bachelors and masters courses on data science have been developed everywhere in the world over the last few years. Universities have found that they must offer studies on data science, because the industry needs experts on it, but also developments in statistics in recent years have almost exclusively come from an area called computational statistics. Statistics is the original form of computing data, and computational statistics takes this to an extreme where methods and tools are developed in a highly data-dependent manner, using and developing modern computational tools. Computational statistics and data science are closely related. Computational statistics covers a broad swathe of data science, exclusive data management, and data security issues. Computational statistics (and therefore also data science) has become very popular since the eighties, and it is very likely the most influential area of statistics nowadays. In the field of computational statistics, not only is new methodology developed, but it is also implemented in software – nowadays almost exclusively in the old but modern software environment R.
Data science seems like a good term when your work is driven by data with a less strong component on method and algorithm development than computational statistics, but with a lot of pure computer science topics related to storing, retrieving, and handling data sets. It also differs from computational statistics in various aspects. For example, in the area of data visualization also pure process-related visualizations (airflows in an engine, for example) are a topic in data science but not in computational statistics.
Wikipedia defines data science as a field that:
incorporates varying elements and builds on techniques and theories from many fields, including math, statistics, data engineering, pattern recognition and learning, advanced computing, visualization, uncertainty modeling, data warehousing, and high performance computing with the goal of extracting meaning from data and creating data products.
Data science is the management of the entire modeling process, from data collection, storage and managing data, data pre-processing (editing, imputation), data analysis, and modeling, to automatized reporting and presenting the results, all in a reproducible manner. It is thus also an interdisciplinary study to extract meaning from data with statistics, by using a lot of elements in computer science, as well as general subject-matter skills. In that sense, data science is an extension and continuation of statistics. Data scientists use statistics and data-oriented computer science tools to solve the problems they face.
Statistical simulation is an essential area in data science. The core issues of this book are simulating distributions and data sets, Monte Carlo methods for inference statistics, and presenting solutions on computer-intense approaches. This book discusses various areas in statistical simulation, random number simulation, resampling, Monte Carlo methods, statistical theory explained by simulation experiments, agent-based microsimulation, and system dynamics. The aim is to put a book into the hands of readers that explains methods, gives advice on the use of those methods, and provides computational tools to solve common problems in statistical simulation and computer-intense methods.
In this book, the theory is not just explained. The theory is also made understandable with illustrative examples using the R software environment. The reader will get to grips with the R software environment. After getting the background on popular methods in the field, readers will see applications in R to better understand the methods, as well as to gain experience when working on real-world data and real-world problems.
R itself is perfectly suited to carry out simulations. It should be mentioned that the basics of R are not the topic of the book, but advanced data manipulation and advanced visualization tools are shown in R. The reader should therefore not be a complete newbie in R, and if so, should first read a very basic introduction to R.
Readers will get a brief overview of the problems and possibilities of data-driven simulation and resampling methods.
What is simulation and where is it applied?
Statistical simulation is a numerical method for conducting experiments on a computer in order to solve mathematical problems in a data-driven manner.
Each experiment is carried out in two steps:
Drawing a random outcome.
The subsequent application of an estimation function to the drawn data.
Random draws are made by means of simulating random numbers, such as the numbers produced by a chosen random number generator.
Simulation is applied in different ways. It is applied in sampling to gather information about a random object by observing many realizations of it (Kroese et al., 2014).
As computational power keeps increasing, and new methods and algorithms are being developed, opportunities present themselves to not only conduct innovative research, but also to design better social and economic policies and programs through micro-simulation and agent-based modeling, where states change over time according to defined transition probabilities. Another example is the area of system dynamics, which describes the interaction of populations or individuals. Both topics are part of Chapter 11, System Dynamics and Agent-Based Models.
With simulation experiments, one can even show the concepts of probability theory and the basic theorems of statistics. The (weak and strong) law of large numbers can be explained. We may repeatedly do some experiments with related probability mechanisms. The outcomes of these experiments are random – random events have outcomes that are not known with certainty, but in the long run we know the properties. To toss coins is the simplest example. The most fundamental theorem in mathematical statistics, the central limit theorem, can also be shown by simulation. Using simulation experiments, readers will be able to fully understand this important theorem, while the proof with mathematics needs very detailed knowledge of mass and probability theory. More information on simulation to show the basics in statistics is provided in Chapter 6, Probability Theory Shown by Simulation.
Statistical simulation is also used to show the properties of an estimation method regarding different conditions. One example is the question of how an estimator behaves under different kinds of missing values pattern, or how outliers may corrupt the estimator. When samples have been drawn with complex sampling designs from finite populations, the influence of the sampling design on the estimator of interest can be shown with design-based simulations. Both model-based simulation and design-based simulation are shown in Chapter 10, Simulation with Complex Data. How data might be simulated for different kind of problems is also discussed. This covers the high-dimensional data and complex synthetic populations needed for design-based simulations.
Usually, when statisticians talk about simulations, they mean Monte Carlo simulations. The Monte Carlo simulation method uses repeated random sampling to mimic the null hypothesis or simulate data from a model where an estimation function is applied to the simulated data.
The Monte Carlo simulation approach is also essential in Bayesian statistics, where Markov chain Monte Carlo (MCMC) methods are used to sample parameter values from a posterior distribution (see also Kroese et al., 2014). This will be intensively discussed in Chapter 4, Simulation of Random Numbers. Generally, it is crucial to have an excellent random number generator at hand that allows you to simulate uniformly distributed values. Additionally, it is crucial to transform a uniform distribution into a distribution of interest. We can do this with inversion, rejection sampling, or MCMC methods; see also Chapter 4, Simulation of Random Numbers.
The Monte Carlo simulation approach is also central to estimating certain numerical quantities in general, but especially to estimate statistical uncertainty. It turns out (Chapter 6, Probability Theory Shown by Simulation) that almost no mathematics is used to express the statistical uncertainty for any complex estimator. The Monte Carlo simulation method is a data-driven and computational tool. It is the perfect tool for data scientists to make statistical inferences without getting lost in the world of mathematics.
Another application of Monte Carlo simulation is multi-dimensional integrals, which can be solved via Monte Carlo techniques, typically by drawing random numbers