Hello!
I am currently working with attributes that influence the visit rate. However, I have identified that some of these attributes might overlap, resulting in collinearity in the data when applying linear regression. My question is: does the model identify this issue? How does it handle collinearity, and what steps does it take to address it?
Thank you in advance.
Alondra
Hi @alcarilu ,
Great question. The model does not identify or avoid collinearity in predictor variables. The only variable transformation the model does it to log-transform the response variable (PUD_YR_AVG). Visitation: Recreation and Tourism — InVEST® documentation
In practice, it is very common to want more control over the regression model. So it is often a good idea to take the data from predictor_data.shp and do data exploration on your own, outside of InVEST. And then use some other statistical modeling software, such as R, to fit your own regression model.