Vol. 10, Issue 12 (2021)

Author(s):
Abhinaya D, Patil SG, Djanaguiraman M and Gunasekaran

Abstract:
The use of principal component analysis in the development of statistical models for crop yield forecasting has been demonstrated. This study employed time series data on groundnut yield and weekly weather variables, such as minimum and maximum temperature, relative humidity, wind speed, and rainfall, from 1990 to 2019 during kharif season in the Erode district of Tamil Nadu. Weekly data on weather variables was used to create weather indices (Agrawal et al., 1983). Four models were created with principal component analysis as Independant variables which also includes time trend and groundnut yield as dependant variable. The model performance was measured using Adjusted R-squared (adj R²) and Root Mean Squared Error (RMSE) as goodness-of-fit criteria. On the basis of adj R² and RMSE, model 1 which includes all the calculated weather indices, was found to be the best suited model with high adj R² (65.51%) and least RMSE (254.7343). Hence, this model can be used to forecast groundnut yield for the studied region.