COMMON MISTEAKS
MISTAKES IN
USING STATISTICS: Spotting and Avoiding Them
Common Mistakes in Regression Related to Model Assumptions
"Lurking
behind the typical regression model will be found a host of such
assumptions; without them, legitimate inferences cannot be drawn from
the model. There are statistical procedures for testing some of these
assumptions. However, the tests often lack the power to detect
substantial failures. Furthermore, model testing may become circular;
breakdowns in assumptions are detected, and the model is redefined to
accommodate. In short, hiding the problems can become am ajor goal of
model building."
David A. Freedman (2010), Statistical Models and Caual Inference,
p. 14
Overfitting
Misinterpreting
the Overall F-Statistic in Regression
Using
confidence intervals when prediction intervals are needed
Over-interpreting
high R2
Mistakes
in interpretation of
coefficients
Mistakes in selecting terms
Further resources concerning cautions in regression:
- R. A. Berk (2004), Regression
Analysis: A Constructive
Critique, Sage
- R. D. Cook and S. Weisberg (1999), Applied Regression
Including
Computing and Graphics, Wiley
- D. A. Freedman (2010), edited by D. Collier, J. S. Sekhon,
and P. B. Stark, Statistical Models
and Causal Inference: A
Dialogue with the Social Sciences, Cambridge University Press.
Hightly recommended.
- P. Good and J. Hardin (2006), Common
Errors in Statistics (and How to Avoid Them), Wiley;
Chapters 10 - 13 and Appendix A
- T. Ryan (2009), Modern
Regression
Methods, Wiley
Last updated April
15, 2011