I know that the title of this question can be misleading but trust me, i didn't know what title to put. I'm doing a project in the field of data science and, before to use any machine learning algorithm, i'm oding independence statistical test to see if there is a relation between the predictors (x_1, x_2,..., x_n) and the target variable Y. I have both continuous and categorical predictors and the target variable is categorical. I already did some independence test to check the level of correlation between the generic x_i and Y :
for continuous predictors -> whitney-mann U rank test (non parametric test)
for categorical variables -> Cramer's-V test (based on the chi-square test)
I know that in my specific case, since i'm going to use decision trees and other stuff related to the filed of machine learning, i could stop doing tests and jump to the next section, but i'd like to know if there are other test that i could perform in order to increase the amount of information that i have on the data. For example:
- Is it possible to test the independence between a paired set of predictors and the target variable (x_1, x_2) <-> Y ? If yes, can i apply it to both continuous and categorical variables or i have to look for other solutions?
Thank you all in advance.