Part II - Multiple regression and diagnostic testing
Published online by Cambridge University Press: 05 June 2014
Summary
In Chapters 2–4, we introduced the basic OLS techniques associated with simple regression analysis (SRA). But, as explained in Chapter 1, OLS is the Best Linear Unbiased Estimator (BLUE) – i.e. the most reliable and accurate estimator of parameters – only when the Gauss–Markov (GM) assumptions are satisfied. In many cases, GM assumptions will be violated and OLS will be inadequately simple; for example the omission of important variables will generate biases in SRA. And there may be other econometric problems further compromising the BLUEness of OLS.
In Part II we introduce multiple regression analysis (MRA), i.e. regressions on models with more than one explanatory variable. We also outline some commonly used econometric tests and procedures, for example the diagnostic tests used to discover violations of the GM assumptions. We will also introduce some procedures for correcting violations of GM assumptions.
- Type
- Chapter
- Information
- Running RegressionsA Practical Guide to Quantitative Research in Economics, Finance and Development Studies, pp. 107 - 108Publisher: Cambridge University PressPrint publication year: 2009