DOI: 10.36347/sjpms

Pages: 196-204

Both exponential smoothing and Box-Jenkins‟ ARIMA models are used in this study as time series modeling approaches to forecast sugarcane variety adoption in Kenya. The accuracy of the two methods are assessed and ARIMA (4,1,1) was found to be the best model to estimate the future prediction of adoption status. Efforts were made to forecast the future adoption of sugarcane variety (KEN 83-737) for two years by fitting ARIMA(4,1,1) model to our time series data. The results indicated a predicted drop in adoption of KEN 83-737 in 2012 and 2013. Keywords: Exponential smoothing, ARIMA, Sugarcane, Forecasting, Time series