[Gretl-users] subsampling and HAC option

Hi I wanted to know how you can get HAC standard errors once you have sub-sampled the dataset, since sub-sampling transforms the initial time-series dataset into a cross-sectional one ? The users guide says you can redefine the structure of the data, but when you choose the initial starting date, the final date does not correspond anymore to what it was initially, due to removed value. Anyway, I could not get the HAC std errors... Thanks for your help!