I am trying to fit the number of crimes in a city with some enviromental variables (aka my features). I'm using a Poisson/Negative Binomial model since I have count data. The problems are:

  1. selecting the features which REALLY fit (I have sometimes big collinearity)
  2. select the smallest subset of significative features (p-values)
  3. understand the relative importance between them to fit the number of crimes

Possible solutions to the problems:

  1. Variance inflation factor (VIF). Is it ok? Sometimes interactions could be ignored. I mean: what if I have to test whenever two variabiles are important, but collinear? (e.g. population density and employment density)
  2. Stepwise (backward/foward/with Cross-Validation) steps to reduce the AIC. However, this does not mean that the resulting variables are significative (p-values). Should I care at all about the features' p-values?
  3. Standardize the input data, then looking at the Beta coefficient to see the importance of them. Is it really the best way?

Should I use something else? Keep in mind that my primary goal is not the $R^2$. I want to see and to be able to explain how some variables could explain the crime.


Firstly, Variance Inflation Factor is inversely proportional to the $R^2$ value. So checking this might help , but multi-collinearity is a thing to look at.

Secondly , instead of $R^2$ , adjusted $R^2$ is a better way to validate a model.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.