EViews.com

Hi

When estimating a ridge equation I was wondering what the L1 and L2 options for regressor transformation do

Thanks

They scale the regressors by the L1 and L2 norms of the data, respectively, prior to estimation. So the penalties on parameters will be on coefficients that are for scaled data. The coefficients are transformed back to original scale for reporting.

Hope that this answers your question.

Where L1 is the sum of the absolute values, and L2 is the square root of the sum of the squares.

yea that makes sense, thanks Gareth