Day 1: Introduction to R Statistical Analysis Software

Qingyin Cai

Department of Applied Economics
University of Minnesota

Summer 2025

Outline

Introduction
- Course Review
- Icebreaker
Motivation
- What you can do with R?
- How do we use R in course in APEC?
- Basic knowledge about R and Rstudio

This one-week course is a boot camp designed to introduce you to R statistical software. My goal is to build a strong foundation for your Ph.D.-level econometrics courses and future research.
By the end of this week, you will be able to:
- to create and manipulate the base-R object data.
- to do data manipulation with data.table package.
- to do data visualization with ggplot2 package.
- to conduct regression analysis with lm() and make a publish-ready regression table with modelsummary() package.
- to write Monte code for Carlo simulations using for loop function.

We will meet each day from 1:00 PM to 4:00 PM, with office hours immediately following.
Each lecture is divided into three sessions, with each session consisting of a 50-minute lecture and a 10-minute break.
We will have in-class exercises at the end of each topic, and after-class exercises (optional) to practice!

No textbook is required. Below are recommended resources:

Recommended Reading

Essential Cheatsheets

Qingyin Cai
- From China
- Fifth-year Ph.D. in Applied Economics
- Area of interests: Food and Agriculture Economics, Consumer Economics, and Environment Economics.

Introduce yourself
- What’s your name?
- What’s your program?
- Where are you from?
- What brings you to UMN?

R is a powerful programming language for a wide range of tasks:
- Data Manipulation: cleaning, reshaping, merging datasets, API.
- Various Analysis: descriptive analysis, regression, GIS, spatial analysis, machine learning.
- Data visualization.
It’s a great tool to communicate your results with others (documentation, papers, slides, books, etc.).

Comparison of R, Stata, and Python

Criteria	R	Stata	Python
Primary Use	Statistical analysis, visualization, research	Economics/social science research; valued for tested results	General-purpose; machine learning, web scraping, automation
Cost	Open-source	Commercial license	Open-source
Data Visualization	Excellent	Less flexible or aesthetically pleasing	Very powerful, but can be verbose
Ecosystem	Large academic community; many packages on CRAN	Strong in economics but smaller user base	Huge, diverse community
Handling Big Data	Base R is memory-bound; packages like `data.table` / `arrow` improve performance	Memory-bound	Excellent

AI can help, but it cannot replace understanding.
- Tools like ChatGPT or Copilot can generate R code, but you need to know if the code is correct and appropriate.
- You’ll understand why a method works, not just how to run it.
Academic integrity & skill development.
- Employers and researchers may expect you to adapt and debug code yourself.
Long-term benefit.
- Once you know R, AI becomes a more powerful assistant — you can ask better questions and spot mistakes.

We use R extensively in the Econometrics series (APEC 8211-8214).
- To conduct regression analysis (OLS, IV, FE, etc.).
- To conduct Monte Carlo simulations.
  - e.g., To understand the difference in variance inference techniques.
Don’t worry if you are new to this!
- Basic knowledge is enough to start.
- If you would like to learn R programming further, I recommend that you take Programming for Econometrics (APEC8221) and Big Data Methods in Economics (APEC8222).

You can use app to write and run R codes, but it has a terrible graphic user interface.
Rstudio is an Integrated Development Environment. It provides a user-friendly interface to write and run R code, view plots, and manage files.
You must install R (the engine) before you can use RStudio!
R studio looks like this:

To create new R script file, click the + button on the top-left corner of the Rstudio, or hit Ctrl + Shift + N (Cmd + Shift + N on mac).
To save the file, click the floppy disk icon , or Ctrl + S (Cmd + S on macOS).

You can change the appearance of Rstudio by going to Tools -> Global Options -> Appearance -> Editor theme and select your favorite theme.

You can have multiple code panes in Rstudio.
To create a new pane, go to Tools -> Global Options -> Pane Layout -> Add Column.
In the same window, you can also change the layout of the panes.

Recent R-studio has a new feature called “Command Palette.”
Hit Ctrl + Shift + P (Cmd + Shift + P on macOS) on your keyboard, or go to Tools -> Show Command Palette.
From here, you can search for and do almost anything: create new files, open projects, etc.

Let’s write some codes.

R code

Any thing you write in the source (or console) pane is regarded as R code.
To run (execute) the code, select the code line, and click the “Run” bottom, or use the shortcut key: Ctrl + Enter (Cmd + Enter on macOS).

Comment block

Any line starting with a # is a comment. R will ignore it. Use comments to leave notes for yourself and others!

You are now familiar with the basics of RStudio. As long as you know how to create, save, and run a script, you are ready for the next lecture.
For more details, see the official RStudio IDE Cheatsheet.
While RStudio is the most popular tool, you can also run R in other editors like Visual Studio Code to run R. Nevertheless, Rstudio is a great starting point to get familiar with R.