STAT 3110 Course Outline

STAT 3110: Applied Regression

Course Outline

Prerequisites:

STAT1220/STAT1221/STAT1222 and MATH 1242/MATH 2120 Course Description: This course offers an introduction into linear regression analysis and emphasizes data analysis by using statistical software, such as R and SAS, and results interpretation. Course topics include fundamental context for linear regression; parameter estimation and inference for linear regression model; model diagnostic; prediction; strategies for building regression models; linear regression model with categorical predictors; and logistic linear regression models (optional). Reference books:

  1. Linear Models with R (2nd edition) by Julian J. Faraway, CRC Press, Taylor & Francis Group
  2. Introduction to Linear Regression Analysis (4th edition) by Douglas C. Montgomery, Elizabeth A. Peck, and G. Geoffrey Vining, Wiley.
Week Topic
1 Syllabus Preliminary data analysis (review concepts about summary statistics and initiate the calculation using a statistical software)
2 Simple Linear Regression Model:

  • Modeling
  • Least square estimation
  • Estimation of variance
3 Simple Linear Regression Model:

  • Confidence interval
  • Prediction interval
4 Simple Linear Regression Model:

  • Hypothesis testing
  • ANOVA and coefficient of determinant
  • A case study
5 A review on matrix algebra and quadratic forms Multiple linear regression model:

  • Modelling and its connection with simple linear regression
  • Parameter estimation
6 Multiple linear regression model:

  • Hypothesis testing

Midterm Exam

7 Multiple linear regression model:

  • ANOVA
  • Confidence intervals
8 Multiple linear regression model:

  • Prediction intervals

Model diagnostics: residual plots

Model diagnostics: QQ plot Unusual points identification: leverage, outlier and influential
9 Model adequacy checking: partial regression plot, collinearity
10 Data transformation and weighted least square
11 Weighted least square
12 Polynomial regression models and categorical predictors
13 Model selection procedures
14 A case study
15 Logistic regression model and review