Intermediate Statistics	27:202:543
Tuesday, 1:00-3:40	Room: CLJ 572
frank.edwards@rutgers.edu
Office hours Thursday, 1:00 - 3:00	Room: CLJ 579B

Quick links

Lecture slides

Homework

Course description

This is the course syllabus for Intermediate Statistics, Spring 2024. Continuous outcomes that meet the assumptions of ordinary least squares regression are relatively rare in the social sciences. This course focuses our attention on how to estimate regression models for discrete outcomes including binary, categorical, and count variables. These flexible tools allow us to more accurately model a wide range of outcomes.

Communication

All course communication will occur over email and canvas. Check them routinely.

Course goals

1. Master data analysis with linear and generalized linear regression models
2. Develop expertise in advanced statistical programming and data visualization
3. Develop the ability to design and conduct quantitative research
4. Become familiar with advanced techniques, like multilevel modeling and missing data analysis

Expectations

• Come prepared. This is a relatively small and advanced course. I expect everyone to participate actively in course discussions.

• Be respectful and professional. Be mindful of the space you take up in the classroom.

• Collaborate with your colleagues. I encourage you all to work together to complete assignments. However, I do expect you each to submit your own homework writeups.

Prerequisites

A prior graduate-level course in statistics is required. This course assumes students are comfortable with multivariate linear regression, basic probability, and statistical computing.

Review resources

These math camp materials from UChicago neatly cover the math you need for graduate-level statistics courses.

Jenny Bryan’s STAT 545 course at UBC provides a very comprehensive overview of programming in R and efficient data science workflows.

Rohan Alexander’s Telling Stories with Data provides a great introduction to practical data analysis with R.

Software

All instruction will be conducted in the R statistical programming language. R is free and open-source, and can be downloaded here.

We will be using the RStudio integrated development environment. RStudio provides a powerful text editor and a range of very useful utilities.

In addition to writing code, it is a great tool for writing reports, papers, and slides using RMarkdown. This syllabus, most of my course materials, and most of my academic papers are based on Markdown. All course assignments should be completed in RMarkdown.

Books (recommended, not required)

These recommended books are very useful, and some examples are pulled from them:

Wickham, R for Data Science

Healy, Data Visualization: A Practical Introduction

McElreath, Statistical Rethinking: A Bayesian Course with Examples in R and Stan

Assignments and grading

Course grading is based on a combination of homework assignments (50 percent) and a final project (50 percent)

Homework

All assignments will be posted and submitted on Canvas

• Problem sets: I will assign weekly homework. These assignments should be completed in RMarkdown and submitted via canvas before class begins the following week.

• Research project: You will write or revise a quantitative paper during the semester.

Course topics and schedule

1/16	Fundamentals: probability, regression
1/23	Inference and simulation
1/30	Linear regression review
2/6	Binary variables and the Bernoulli distribution
2/13	Linear regression (1)
2/20	Logistic regression (2)
2/27	Models for count data (1)
3/5	Models for count data (2)
3/12	Spring Break
3/19	Models for categorical outcomes
3/26	Advanced topics: missing data (1)
4/2	Advanced topics: missing data (2)
4/9	Advanced topics: multilevel models (1)
4/16	Advanced topics: multilevel models (2)
4/23	Advanced topics: Bayesian regression