2

The covTest package in R gives significance values for a LASSO regression. How should the results be interpreted? I get negative predictor numbers and NAs for the p-values. What do these mean? More specifically, it might be assumed that the predictor number is an index for the external regressors, but how can this be negative?

Here is an example:

 significance_pars <- covTest(lars_output,x,y, sigma.est = "full")

When calling significance_pars$results, the output is:

 Predictor_Number Drop_in_covariance P-value
          118          8830.0759  0.0000
          105          1557.1987  0.0000
          104           755.5005  0.0000
          119           833.6093  0.0000
           46           183.8750  0.0000
           45             3.3674  0.0345
          103            56.8601  0.0000
           60           214.5187  0.0000
           44             0.4365  0.6463
          113           248.4377  0.0000
          120           220.6486  0.0000
          124             1.2863  0.2763
         -118                 NA      NA
fifthace
  • 121

0 Answers0