Course Overview

Computational data analysis is an essential part of modern statistics. This course is designed to help students develop programming skills and new computational tools for real data analysis. Through this course, students will learn the core ideas of statistically-oriented programming and how to debug and test their code. This course also covers the design of stochastic simulations experiment, statistical model fitting, large data sets manipulation and basic data visualization. This course will make a balance between programming skill training and real data analysis with examples. We will make use of the statistical computing software R in class and on homework assignments. Extensive R programming background is not required.

Textbook

• The Art of R Programming: A Tour of Statistical Software Design, by Norman Matloff (Required)

• Statistical Computing with R, by Maria L. Rizzo (Required)

• R for Data Science, by Hadley Wickham & Garrett Grolemund (Required)

• The R Cookbook, by Paul Teetor (Optional)

• The R Graphics Cookbook, by Winston Chang (Optional)

• R Markdown Cookbook, by Yihui Xie, Christophe Dervieux, Emily Riederer (Optional)

Topic Outline

1. R and data structure

2. R Markdown, indexing and iteration

3. Data frame and vectorization

4. Functions and objective-design

5. Testing and Debugging

6. Plotting

7. Simulation

8. Optimization and numerical methods

9. Read and write data in R

10. Fit the model

11. Tidyverse and versional control

12. Make own R package

Software

• R and RStudio (preferred) will be used and all are free to download. (Getting started with R and RStudio)

Grading

• Weekly quiz (20%): The lowest score will be dropped.

• Homework assignments (30%): There are totally 4 assignments.

• Final project (50%) : Presentation (20%) and report (30%)

More reading materials

• Quick-R

• R code for examples from the "Statistical Computing with R"

• Hands-On Programming with R

• Data Visualization with R

• ggplot2: Elegant Graphics for Data Analysis