Song Naiping, Zhang Rongqun, Lv Bo, Ai Dong. Predicting Ningxia Province Total Grain Yield Using Univariate Methods and Trend Models. Sensor Letters, 2014, Vol.12,1–5.
The forecasting of total grain yield can help to improve decision making in agricultural production management. In this paper, a grey-Markov chain forecast model for total grain yield is presented. The purpose of this work was to seek a combined method based on the grey model and the Markov chain model in order to provide the most accurate prediction of a univariate time series. From annual total-grain-yield data for the period 1985–2006 for Ningxia Province, China, we obtained new timeseries data using GM (1, 1). The time series annual data was divided into several interval states based on the relative error between the actual value and the GM (1, 1) prediction, and the GM (1, 1) prediction correction coefficient was computed using the Markov chain model. Prediction results are assessed based on mean absolute error, mean absolute percentage error and root mean squared error. The grey-Markov chain forecast model improves greatly the accuracy of the predicted values of the total grain yield for Ningxia Province.