I want to see what effect multi-collinearity has on a linear regression model but I need to be able to generate multi collinear data where I can vary the number of features and the collinearity between these features.
I've had a look at Sklearn's make_regression
function and it allows for the generation of multiple features but from what I understand these features are all uncorrelated correct?
If so, does anyone know how I could vary the correlation between these features or use a different method to generate a linearly multi-collinear dataset to train Sklearn's linear regression model with?