STAT 230

STAT 230 - Linear Models

Class meetings

Mondays and Thursdays

10:00 AM - 11:30 AM

Building 9, Room 4229

Lecturer

Dr. Paula Moraga

E-mail: paula.moraga@kaust.edu.sa

Office location: Building 1, Al-Khwarizmi, Office 4302

Website: https://www.paulamoraga.com/

Teaching Assistant

Mr. Jesus Diaz Martinez

E-mail: jesus.diazmartinez@kaust.edu.sa

Course materials

https://www.paulamoraga.com/book-stats

Course overview

This course is an introduction to the formulation and use of linear models and generalizations including parameter estimation and inference for such models in a variety of settings. Emphasis will be split between understanding the theoretical foundations of the models and the ability to apply the models to answer scientific questions.

Statistical analyses will be conducted using R and the integrated development environment (IDE) RStudio.

Example: How does physical activity relate to body mass index (kg/m^2)?

Course schedule

Week 1: Statistical inference foundations (probability distributions, central limit theorem)
Week 2: Confidence intervals, hypothesis testing, ANOVA
Week 3: Simple and multiple linear regression (least squares, geometric interpretation, confidence intervals, hypothesis testing)
Week 4: Multiple linear regression: Categorical variables
Week 5: Saudi National Day. Multiple linear regression: Interactions
Week 6: Model assumptions and unusual observations. Residuals and leverages
Week 7: Generalized and weighted linear squares. Model selection
Week 8: Mid-semester break. Midterm exam
Week 9: Generalized linear models (GLMs), Poisson regression, logistic regression
Week 10: Generalized linear models (GLMs), exponential family, deviance, inference
Week 11: Classification, ROC curve
Week 12: Survival models: censorship, Kaplan-Meier, Cox proportional hazards model
Week 13: Statistics Workshop
Week 14: Local regression, splines. Generalized additive models (GAMs)
Week 15: Spatial models
Week 16: Final exam

Assignments

There will be homeworks that will be submitted to Blackboard by the due date. Late homeworks will not be accepted, unless prior arrangements have been made.

Exams

There will be one midterm exam to be scheduled inside of the regular class hours, and one comprehensive final exam to be scheduled by the registrar’s office.

No written material may be brought into the examination, except for one page (one side) of handwritten notes. A simple calculator may be used.

Midterm exam on Thursday 17th October.

Example Midterm exam

Method of evaluation

50% - Final exam
25% - Midterm exam
25% - Homework

Required knowledge

Advanced and multivariate calculus, linear algebra, probability and statistics.

References

1. Christensen (2011) Plane Answers to Complex Questions: the Theory of Linear Models, Springer
   
2. Wood (2015) Core Statistics, Cambridge University Press
   
3. Seber and Lee (2003) Linear Regression Analysis, Wiley
   
4. Hocking (1996) Methods and Applications of Linear Models: Regression and the Analysis of Variance, Wiley
   
5. McCullagh and Nelder (1989) Generalized Linear Models, Chapman & Hall/CRC
   
6. Kariya and Kurata (2004) Generalized Least Squares, Wiley
   
7. Hastie and Tibshirani (1990) Generalized Additive Models, Chapman & Hall/CRC
   
8. Davison (2003) Statistical Models, Cambridge University Press
   
9. Faraway (2005) Linear Models with R, Chapman & Hall/CRC
   
10. Faraway (2006) Extending the Linear Model with R, Chapman & Hall/CRC
    
11. Wood (2006) Generalized Additive Models: An Introduction with R (Chapman & Hall/CRC)
12. Moore (2016) Applied Survival Analysis Using R. Springer

License

You may not copy or distribute the course materials.