STAT 301 Lecture Notes - Dependent And Independent Variables
Document Summary
As we add predictors to a model, the r2 for the model will increase regardless of whether the added predictor is useful. Cannot compare useful r2"s if # of predictors is different adj was developed to allow such a comparison. R2 adj = 1 [ n 1 n (k+1) Tss ] n = sample size k is number of predictors. In theory, we choose the model with the highest r2 adj. In practice, we find a group of models with similar r2 with the fewest predictors) adj and choose the simplest (the one. Sometimes the change in y associated with a 1-unit increase in one explanatory variable depends on the value of a second explanatory variable. Change in y associated with x1x2, an interaction. If x1 and x2 interact, our model is. Y = 0+ 1x1 + 2x2 + 3x1x2 + .