Stat 505 - Linear Models

STAT 505 is the first semester in a two semester sequence of courses designed to help you gain a deeper level of understanding of the most common statistical methods used by statisticians. STAT 505 is a course in the theory and application of linear models, the foundation of most models used in statistical analysis. Topics include: Special matrix theory for statistics, multivariate normal distribution, distributions of quadratic forms, estimation and testing for the general linear model, one-way and two-way classification models, contrasts (main effect, simple effect and interaction), and multiple comparison techniques.

By the end of this course, the successful student should be able to:

1. Use rigorous mathematical techniques and methods of proof to derive results in matrix theory for statistics.
2. Derive estimators, hypothesis tests, and confidence intervals related to linear models.
3. Understand the assumptions, uses, and limitations of linear models.
4. Use R to fit linear models to data and conduct inference on linear model parameters, with and without using built-in functions.
5. Communicate (written, visually, and orally) results of a statistical analysis and statistical concepts in non-technical terms.

Course Information

Log into D2L course page to access course material, assignments, etc.

• Syllabus
• Course calendar

Data Sets

• Happy faces and restaurant tips (Example 16.15 in Utts and Heckard, 5th ed.)
• Cloth flammability tests (from Elementary Statistics, 7th ed. by Triola)
• IPEDS data Excel file (Source: https://nces.ed.gov/ipeds/use-the-data)

R References

• R and RStudio basics
• Quick-R
• A Beginner's Guide to R
• R cheatsheets (see "Data visualization with ggplot2" and contributed cheatsheet "Base R")
• Datacamp's free Introduction to R Programming online course
• Guru99's free R Tutorial for Beginners 

Links to additional R Resources are posted in our D2L Content page.