Skip to main content Accessibility help
×
Home
  • Print publication year: 2010
  • Online publication date: June 2012

6 - Diagnostic testing

Summary

Learning outcomes

In this chapter, you will learn how to

describe the steps involved in testing regression residuals for heteroscedasticity and autocorrelation;

explain the impact of heteroscedasticity or autocorrelation on the optimality of OLS parameter and standard error estimation;

distinguish between the Durbin–Watson and Breusch–Godfrey tests for autocorrelation;

highlight the advantages and disadvantages of dynamic models;

test for whether the functional form of the model employed is appropriate;

determine whether the residual distribution from a regression differs significantly from normality;

investigate whether the model parameters are stable; and

appraise different philosophies of how to build an econometric model.

Introduction

Chapters 4 and 5 introduced the classical linear regression model and discussed key statistics in the estimation of bivariate and multiple regression models. The reader will have begun to build knowledge about assessing the goodness of fit and robustness of a regression model. This chapter continues the discussion of model adequacy by examining diagnostic tests that will help the real estate analyst to determine how reliable the model is and to recognise the circumstances under which OLS may run into problems. These tests enable an assessment of the quality of a model, the selection between models and, in particular, an assessment of the suitability of the chosen model to be used for forecasting.

Related content

Powered by UNSILO