I want to perform a regression where my independent variables all sum to 1. The independent variables are proportions of money invested in different categories.
What I have done:
- When checking the vif of the regression I get an error that 'there are aliased coefficients in the model', which means I have a multicolinearity problem.
- I remove one of the independent variables and after running the vif, there is no problem any more (all values < 5). I know vif is not the only way to test multicolinearity.
- I have checked for autocorrelation and I do not have that problem.
Is this enough? Should I be running another type of regression instead? I am no expert, which is why I do not know if what I have done is enough.
(I will remove it if not appropriate)
- Edit: I added an image with sample of my data (the column missing is the last Industry that has the rest of the proportion invested so that each column adds to 100%).
