การพยากรณ์ราคาข้าวส่งออกของประเทศไทยโดยใช้แบบจำลอง ARIMAX

Abstract

The purposes of this study were to find out suitable model; and forecasting rice price for 4 types include Thai Hom Mali Rice 100% class 1 (JassA), Thai Hom Mali Rice class 2 (JassB) and Thai White Rice 100% class 1 (WRA), Thai White Rice 100% class 2 (WRB) export of Thailand using currency exchange rates USD/Baht and World rice prices as extraneous variable for rice export pricing of Thailand. Data was collected, started from January 2546 to December 2558 for 156 examples. From testing found unit roots in seasonal time series found prices of Thai Hom Mali Rice 100% class 1 (JassA), Thai Hom Mali Rice class 2 (JassB) and Thai White Rice 100% class 1 (WRA), Thai White Rice 100% class 2 (WRB) currency exchange rates USD/Baht and World rice prices do not have seasonal unit root but do have yearly unit root by test from T-test and F-statistic were statistically significant at the 0.05 level compared with critical value of franses at the 0.05 level. 2559 using ARIMAX found suitable model and ex-ante forecast period were as follow : Thai Hom mali 100% (JassA) using ARIMAX model (1,1,0) forecasts equal to 809.06, 807.86, 809.68 ( US$/MT), Thai Hom mali 100% (JassB) using ARIMAX model (1,1,0) forecasts equal to 780.61,779.55,781.37 ( US$/MT), Thai white rice 100% (WRA) using ARIMAX model (1,1,1) forecasts equal to 392.93, 393.46, 394.56 (US$/MT), Thai white rice 100% (WRB) using ARIMAX model (1,1,1) forecasts equal to 373.31, 372.31, 374.49 ( US$/MT). Therefore, concluded that currency exchange rates USD/Baht were statistically significant at the 0.05 level for Thai Hom Mali Rice 100% class 1 (JassA), Thai Hom Mali Rice class 2 (JassB) but do have Thai White Rice 100% class 1 (WRA), Thai White Rice 100% class 2 (WRB) and World rice prices do not have statistically significant at the 0.05 level with Thai rice price include 4 types.