Prediction of Cow Milk Yield in Himachal Pradesh and Northern Himalayan Province of India

Author: Rahul Thakur, Ashu Chandel, R.K. Gupta, Rohit Bishist, Vaibhav Chittora* and Pawan Kumar

Download PDF

Abstract

Animals are an integral part of our farming system. Livestock production plays a major economic as well as cultural role in the rural community. It provides an indirect means of insurance against the risk of crop failure due to natural calamities such as drought and flood. This paper makes an attempt to predict the yield of cow milk in Himachal Pradesh Northern Himalayan state of India so that they can enhance the production of cow milk. Secondary data on cow milk production from 1993 to 2014 (22 years) of Himachal Pradesh were used for the purpose and different linear and non linear models were applied. The data were subjected to regression analysis and trend value of milk yield was also assessed. Various models such as Linear, Quadratic, Cubic and Compound models were fitted to predict the cow milk yield. (R^2 ) ̅, RMSE and Theil’s U statistic were used for selection of model. Quadratic model turned out to be best fit for cow milk yield. Validation using F- Chow Statistics was ca

Keywords

Statistical model, linear model, quadratic model, RMSE, adjusted R2, F-chow, Autoregressive model.

Conclusion

In the present investigation linear, quadratic, cubic and compound models were fitted for estimation cow milk production, using 22 years of data. Among these the quadratic model was found best for estimation milk production of cow on the basis of the lowest RMSE and Theil’s U statistics values. Non-significant F values in the Chow test revealed the validity of quadratic model for prediction. Autoregressive models till the 4th order were also fitted to predict milk yield of cow. Both 1st and 3rd order AR models were found to be well fitted and valid for the given year but the 1st order AR model was preferred to the 3rd order model for prediction purpose as per lower value of RMSE. The study further showed that quadratic model had lower RMSE value as compared to the 1st order AR model. Hence, the quadratic model was found to be the best fit and valid model for estimation of cow milk production in Himachal Pradesh, the northern Himalayan state of India. The same may be demonstrated by u

References

-

How to cite this article

Rahul Thakur, Ashu Chandel, R.K. Gupta, Rohit Bishist, Vaibhav Chittora and Pawan Kumar (2022). Prediction of Cow Milk Yield in Himachal Pradesh and Northern Himalayan Province of India. Biological Forum – An International Journal, 14(1): 1078-1082.