Forecasting of wheat (riticum aestivum) yield using ordinal logistic regression

Abstract

In this study, uses of ordinal logistic model based on weather data has been attempted for forecasting wheat (Triticum aestivum L.) yield in Kanpur district of Uttar Pradesh. Weekly weather data (1971-72 to 2009-10) on maximum temperature, minimum temperature, morning relative humidity, evening relative humidity and rainfall for 16 weeks of the crop cultivation along with the yield data of wheat crop have been considered in the study. Crop years were divided into two and three groups based on the detrended yield. Yield forecast models have been developed using probabilities obtained through ordinal logistic regression along with year as regressors for different weeks. Data from 1971-72 to 2006-07 have been utilized for model fitting and subsequent three years (2007-08 to 2009-10) were used for the validation of the model. Evaluation of the performance of the models developed at different weeks has been done by Adj R2, PRESS (Predicted error sums of squares) and number of misclassifications. Evaluation of the forecasts were done by RMSE (Root mean square error) and MAPE (Mean absolute percentage error) of forecast.