Principle Component Analysis for Forecasting of Pre-harvest Rapeseed and Mustard Yield Based on Meteorological Parameters
Sarvesh Kumar *
Department of Agricultural Statistics, Acharya Narendra Deva University of Agriculture and Technology, Kumarganj, Ayodhya, U.P., India.
K.K. Mourya
Crop Research Station, Masodha, Acharya Narendra Deva University of Agriculture and Technology, Kumarganj, Ayodhya, U.P., India.
R.P. Gupta
Department of Agricultural Statistics, Acharya Narendra Deva University of Agriculture and Technology, Kumarganj, Ayodhya, U.P., India.
B.V.S. Sisodia
Department of Agricultural Statistics, Acharya Narendra Deva University of Agriculture and Technology, Kumarganj, Ayodhya, U.P., India.
*Author to whom correspondence should be addressed.
Abstract
The pre-harvest forecast model was developed using time series data on Rapeseed and Mustard yields as well as weekly data on six weather variables from the 40th SMW of one year to the 8th SMW of the next year, which covers the years 1990–1991 to 2014–2015. Multiple regression and principal component analysis are two statistical techniques that have been reported for creating pre-harvest forecast models. The most accurate model found by applying step-by-step regression analysis to weekly weather data was based on adj R2, RMSE, and%SE. One and a half months prior to harvest, these models can be utilised to obtain a trustworthy forecast of the yield of Rapeseed and Mustard.
Keywords: Meteorological parameters, crop yield, multiple regression principle component function analysis, forecast model