STAT151 Chapter Notes - Chapter 8: Confounding, Linear Regression, Subset

• School: University of Alberta
• Department: Statistics
• Course: STAT151
• Professor: Omar Rivasplata

Document Summary

It is a good idea to look at both a histogram of the residuals and a scaterplot of the residuals vs. the predicted values. All the data must come from the same populaion (when more than on group, report two regressions) Can strongly inluence the regression, so be careful! When the x-value is far from the mean of the x-values (this is unusual). It has the potenial to change the regression. A point is inluenial is omiing it from the analysis gives a very diferent model. When venturing into new x-territory, it requires cauion; you are beter of not doing this. No mater what, there is no way to conclude from a regression alone that one variable causes another (possibility of third lurking variable, driving both variables you have observed) There is no way to be sure that a lurking variable is not the cause of any apparent associaion.