ARIMA and General Regression Neural Network for Forecasting Rice Production in Sri Lanka

• Activity Number: 444
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Manjari Dissanayake*+ and Ferry Butar Butar
• Companies: and SHSU
• Keywords: Rice Production ; ARIMA ; GRNN ; Time Series ; K-Medoid

Abstract

Forecasting in food crop has become one of the crucial agricultural factors nowadays. With the help of the modern technology and statistical tools scientists try to evaluate and forecast the crop production, minimizing the errors as much as they can. Time series techniques served the purpose of forecasting for many years by fitting in classical models such as ARIMA, considering the past behavior of the data that already exist. General Regression Neural Network (GRNN) is one of the promising and upcoming areas of research in Statistics at present. It is a special tool to predict and compare system performance in practice. In this paper we discuss and compare the forecasts of rice production using the method ARIMA with the introducing tool GRNN.