High Impact Factor : 4.396 icon | Submit Manuscript Online icon | UGC Approved icon

Study of Effect of Climatological Variables on Crop Yeild Estimation Using Multiple Linear Regression


Dr. T. M. V. Suryanarayana , WREMI, The M.S. University of Baroda


Climatological Data, Crop Yield, Multiple Linear Regression, R.M.S.E., Coefficient of Correlation


An attempt has been made to carry out the study of determining the predominant climatological variables in estimating the crop yield. The climatological data are collected for the period 1981- 2006 and correlated with yield of cotton in Vallabh Vidyanagar using Multiple Linear Regression. The Climatological variables considered are Maximum Temperature, Minimum Temperature, Relative Humidity, Wind Speed and Sunshine Hours. The multiple linear models have been developed, to study their impact in prediction of the crop yield. The study has been carried out with eight different combinations of the five independent variables considered, to correlate with the crop yield. In each combination, i.e 1 to 8, the whole data is divided into proportions for training and Validation, such as 70% and 30% & 60% and 40% respectively. The developed Multiple Linear Regression Models are evaluated based on the performance indices such as Root Mean Squared Error and Correlation Coefficient. Based on the evaluation, the models developed are found to perform better in 60%-40% proportion of the data considered for the Study. Therefore in this considered proportion of the dataset, the models developed are ranked based on the obtained R.M.S.E. and R. The results clearly show that the consideration of all the variables, yield the best model with minimum R.M.S.E. and maximum R, followed by the combinations considering Maximum Temperature, Minimum Temperature, Relative Humidity as dependent variables along with/without Wind Speed/Sunshine hours. Moreover excluding the Relative Humidity, and trying the combinations of Maximum Temperature, Minimum Temperature along with/without Wind Speed/Sunshine Hours yields the poor models with maximum R.M.S.E. amd Minimum R. Hence considering multiple linear regression models and the eight combinations studied, it reveals that the yield of a crop is very much dependent on maximum and minimum temperatures & relative humidity.

Other Details

Paper ID: SPDM021
Published in: Volume : 1, Issue : 2
Publication Date: 01/11/2015
Page(s): 24-24

Article Preview

Download Article