I have ran a step-wise multi linear regression in SPSS.
3 predictor variables - 2 predictors were excluded
average leverage equation (k + 1/n) (where k is the no. of predictors)
Do I calculate using only 1 predictor variable or 3 predictor variables?
i.e.,
1 + 1/n
or
3 +1/n
For multiple regression on n cases, with k predictor variables in the final model, the average leverage value is always (k+1)/n, provided that the final model includes an intercept term. So if your final model included just one predictor variable, it doesn't matter how you got to the one-variable model, the average leverage is (1+1)/n = 2/n
For multiple regression on n cases, with k predictor variables in the final model, the average leverage value is always (k+1)/n, provided that the final model includes an intercept term. So if your final model included just one predictor variable, it doesn't matter how you got to the one-variable model, the average leverage is (1+1)/n = 2/n.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA