10
$\begingroup$

I have recently graduated and want to review some stats topics to prepare for interviews. I was wondering whether there is a good book or some other resource to review linear regression. E.g., I would want to review all the assumptions underlying OLS, and what to do in case an assumption does not hold in practice.

Example: let's say the residuals didn't have an expected mean of zero, what could be the issue? How would you fix it?

I would like to read up on all this stuff again, but do not really know what the most efficient way is to do so. Does anyone know a document or an easy to go through book to review all this stuff?

$\endgroup$
1
  • 1
    $\begingroup$ +1. Surprisingly, a focused site search indicates this question has neither been asked nor answered here before. $\endgroup$
    – whuber
    Commented Apr 11, 2020 at 11:39

4 Answers 4

6
$\begingroup$

This is a comprehensive classic:

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.).

$\endgroup$
6
$\begingroup$

I’m not sure if this answers your needs, but the best book to “get into” regression modeling that I can recommend is

Data Analysis Using Regression and Multilevel/Hierarchical Models by Andrew Gelman and Jennifer Hill

It’s great to develop understanding of regression models and for learning how to apply them. It may not go that deep into the technical details that may come up in interviews, but I’d definitely recommend it as first read before going into more technical texts.

$\endgroup$
3
$\begingroup$

An Introduction to Statistical Learning has an excellent review of simple and multiple linear regression, with about 60 pages dedicated to those topics.

$\endgroup$
2
$\begingroup$

I like Exegeses on Linear Models by W N Venables, old but I still find it provocative.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.