Generalized Extreme Value Regression for Prediction of Meat Import Price Range in Indonesia

Price range is one of the mechanisms that used to detect undervalued or overvalued prices for imported goods. The problem encountered in forming price range is that the data on the price of imported goods has an asymmetric distribution with non-stationary conditions or trends and covariates that change over time. GEV Regression model is a statistical analysis method used to model extreme value data with non-stationary conditions or the presence of trends and the influence of covariates that change over time. This research aims to apply the GEV Regression model to predict the import price range and obtain its significant predictor variables. Next, the ARDL model obtains the proper lag of the predictor variable within the GEV regression. The significant predictor for the maximum price of imported meat is the exchange rate one and two months earlier, along with the maximum price at three and four months earlier. Meanwhile, the significant predictor for the minimum price of imported meat is the exchange rate one month earlier and the minimum price one and two months earlier.