Book Image

Simulation for Data Science with R

By : Matthias Templ
Book Image

Simulation for Data Science with R

By: Matthias Templ

Overview of this book

Data Science with R aims to teach you how to begin performing data science tasks by taking advantage of Rs powerful ecosystem of packages. R being the most widely used programming language when used with data science can be a powerful combination to solve complexities involved with varied data sets in the real world. The book will provide a computational and methodological framework for statistical simulation to the users. Through this book, you will get in grips with the software environment R. After getting to know the background of popular methods in the area of computational statistics, you will see some applications in R to better understand the methods as well as gaining experience of working with real-world data and real-world problems. This book helps uncover the large-scale patterns in complex systems where interdependencies and variation are critical. An effective simulation is driven by data generating processes that accurately reflect real physical populations. You will learn how to plan and structure a simulation project to aid in the decision-making process as well as the presentation of results. By the end of this book, you reader will get in touch with the software environment R. After getting background on popular methods in the area, you 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.
Table of Contents (18 chapters)
Simulation for Data Science with R
Credits
About the Author
About the Reviewer
www.PacktPub.com
Preface
Index

Proper variance estimation with missing values


Very often in practice, missing values are a major problem. Standard routines for estimation are typically not designed to deal with missing values. In the following we discuss a method to adequately deal with missing values when estimating the variance/uncertainty of an estimator.

Because of non-answered questions or measurement errors, data often has the following data structure:

Here we see n observations and p variables and some missing values (NA).

Often one will omit those observations that include missing values from the data set. However, this decreases the sample size and thus increases the variance of estimators, and in addition this may cause biased estimates if missing values are missing at random, that is; if the probability of missingness depends on covariates.

To work around this problem, another, better solution is to impute missing values. For some applications the imputations are done in a way to minimize a prediction error. For...