Use of ARIMA modeling for forecasting green gram prices for Maharashtra Chaudhari D.J.*, Tingre A.S. Department of Agricultural Economics & Statistics, Dr. Panjabrao Deshmukh Krishi Vidyapeeth, Akola-444 104, India *E-mail: datta1616@rediffmail.com
Online published on 30 July, 2014. Abstract The present study aimed to forecast the Green gram prices for Maharashtra by using the time series data of monthly average prices for the period from January 2001 to September 2012 of Akola market. To forecast the Green gram prices ARIMA models introduced by Box and Jenkins (1970) were used. To test the reliability of model R2, Mean Absolute Percentage Error (MAPE), and Bayesian Information Criterion (BIC) were used. Model parameters were estimated using the Statistical Package for Social Sciences (SPSS). Among the different models lowest BIC value worked out to 1989 for ARIMA (0,1,0) which was the best fitted model. Based on model results the estimated Green gram prices for Maharashtra would increasing from Rs. 4646 per quintal during October 2012 to Rs. 4729 per quintal during February 2013. Top Keywords ACF, ARIMA, Auto regression, Box and Jenkins, forecasting, moving average, PACF. Top |