L. Hidayat, D.U. Wutsqa
The transfer function model is one of the forecasting models that predicts future values from a time series based on its past values and the other variables. In this paper, the transfer function model is utilized to forecast the farmers' terms of trade in Yogyakarta province. The farmers' term of trade (FTT) is one indicator of farmers' welfare. The variables included in the model inputs are the inflation and exchange rate in Yogyakarta. The procedure for transfer function modeling is through two stages, those are building a single input transfer function model for each input based on the autoregressive integrated moving average model, and then building the transfer functions simultaneously for all variables. The result demonstrates that the transfer function can predict farmers' terms of trade accurately with the mean absolute percentage error value of 0.841%. © Published under licence by IOP Publishing Ltd.
Department of Mathematics, Faculty of Mathematics and Natural Science, Universitas Negeri Yogyakarta, Yogyakarta, Indonesia; Department of Statistics, Faculty of Mathematics and Natural Science, Universitas Negeri Yogyakarta, Yogyakarta, Indonesia
