Re: [Gretl-users] LAD rolling regression