Model Regresi Linear Produksi Padi di Indonesia dengan Estimasi-M

Abstract

Policy of the price of main food is one of important instrument in creating resilience of national food. Remembering the importance of accomplishment e®ort of requirement of food, especially paddy, it is required an e®ort to predict production in future. There are some methods which are applicable to predict the production of paddy and to investigate the factors in°uencing it; one of them is regression analysis. An estimation method which is applicable to determine the regression model is M-estimation. This estimation is an extension of maximum likelihood method and robust estimation, where its estimation value is not in°uenced by small change in data. The purpose of
this research is to determine the regression model to predict the production of paddy in Indonesia using estimation M.

Based on the result of research we conclude that the prediction model for production of paddy in Indonesia with M-estimation is
by = ¡92; 790 + 5; 02x1 + 6; 37x2:
The increment of one hectare farm wide and one ton seed will increase production of paddy 5.03 tons and 6.23 tons respectively. The regression parameter signi¯cance test shows that the farm wide in°uences the production of paddy signi¯cantly.