I'm trying to determine a quantitative value by which a target variable change (inflation) by changing an indicator variable (interest rate).
The industry basically uses linear models such as VAR. Are there state of the art approaches to capture such relationship?
Yes there are multiple ways to check the relationship between the input and target variables. Some of which are:
1.) Correlation based techniques - You can use Pearson, Spearman or Kendall correlation depending on whether your variables are linearly or non linearly related.
2.) Use mutual_info_gain to check the dependency between 2 variables.
3.) Calculate the feature importance using machine learning models.
There are other techniques you can use too. Google is your best friend!
Cheers!
