I am currently comparing multiple multiple linear regression models to examine the impact of living environments on happiness levels in different regions. However, one of the models exhibits a severe issue of multicollinearity in its samples. Therefore, I have had to employ ridge regression to mitigate the multicollinearity in that particular model.
Now, the question arises: should I convert all of the linear regression models to ridge regression, or should I only apply ridge regression to the model with strong multicollinearity? After all, ridge regression can result in some loss of information. However, is it feasible to compare ridge regression models and linear regression models together if I don't convert all of the models?
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ornithology
- birds
- Scopus
More Ningbo Kou's questions See All
Similar questions and discussions