Case Detail

Case Summary

Practical Regression: Noise, Heteroskedasticity, and Grouped Data

Case Number: 7-112-006, Year Published: 2012

HBS Number: KEL640

Request PreviewBuy

Authors: David Dranove

Key Concepts

Economics, Market Research, Statistical Methods

Abstract

This is the sixth in a series of lecture notes which, if tied together into a textbook, might be entitled “Practical Regression.” The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title “Practical Regression” suggests, these notes are a guide to performing regression in practice.
This note begins by describing the problems that arise when variables are measured with noise. The note progresses to explain how to test for and correct heteroskedasticity. It also explains why it is necessary to adjust standard errors when working with grouped data.

Learning Objectives

Students will learn the following:
-The problems that arise when the dependent and/or predictor variable is measured with noise
-How to diagnose heteroskedasticity graphically and by performing the Breusch-Pagan test
-Approaches for correcting heteroskedasticity, including estimating robust standard errors and performing weighted least squares regression (log specifications are covered in a different note)
-A step-by-step guide for coping with heteroskedasticity

Number of Pages: 15

Extended Case Information

Teaching Areas: Finance