Introduction to Regression Models and Analysis of Variance


Stanford School of Engineering

Introduction to Regression Models and Analysis of Variance


Regression modeling, when used with understanding and care, is one of the most widely useful and powerful tools in the data analyst’s arsenal. This course aims to build both an understanding and facility with the ideas and methods of regression for both observational and experimental data. You will develop competency in choosing the right set of data analysis tools depending on the nature of data along with their limitations.

Please note that is an online-only course, lectures will not be recorded on-campus.

What you will learn

  • How to study the relationship between variables
  • The assumptions underlying regression analysis
  • How to distinguish between random and fixed variables
  • How to choose the right statistical model for the data


  • Prathapa Kaluwa Devage


  • A post-calculus introductory probability course, e.g. Stanford Course STATS116
  • In addition, a co-requisite post-calculus mathematical statistics course, e.g. Stanford Course STATS 200,
  • Basic computer programming knowledge
  • Familiarity with matrix algebra

Topics include

  • Introduction, summary statistics and simple liner regression
  • Multiple regression: geometric characterization, goodness of fit, unbiasdness
  • Inferences, t and F tests
  • Confidence intervals for prediction
  • Regression diagnostic
  • Variable selection
  • Shrinkage methods
  • ANOVA: regression based analysis
  • Modeling and interpretation of data