Re: [Gretl-users] forecasting with exponential smoothing