HYBRID TIME SERIES MODELLING FOR FORECASTING THE PRICE OF WASHED COFFEE (ARABICA PLANTATION COFFEE) IN INDIA

K. NAVEENA¹*, SUBEDAR SINGH², SANTOSHA RATHOD³, ABHISHEK SINGH^4
1Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, 221005, Uttar Pradesh
²Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, 221005, Uttar Pradesh
³ICAR-Indian Agricultural Statistics Research Institute, New Delhi, 110012
⁴Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, 221005, Uttar Pradesh
* Corresponding Author : naveenkdec@gmail.com

Received : 06-12-2016     Accepted : 21-02-2017     Published : 28-02-2017
Volume : 9     Issue : 10       Pages : 4004 - 4007
Int J Agr Sci 9.10 (2017):4004-4007

Keywords : Arabica Plantation Coffee, ARIMA, ANN, Hybrid ARIMA-ANN, Forecasting
Conflict of Interest : None declared
Acknowledgements/Funding : I thank Prof. Subedar Singh for assistance with my Ph.D thesis and Abishek Singh, Asst. Prof of BHU for comments that greatly improved the manuscript
Author Contribution : Naveena K: Original author for this research work

Indian coffee possesses an exotic full-bodied taste and a fine aroma. Indian coffee has created a niche for itself in the international market. Arabica plantation coffee is one of the important types of coffee produced in India. In this study ARIMA, ANN and Hybrid ARIMA-ANN models were used to analyse the past behaviour of a time series data, in order to make inferences about its future behaviour for Arabica Plantation species of Indian coffee. Finally, the forecasting performance of these models are evaluated and compared by using common criteria’s such as; Root Mean Square Error, Mean Absolute Percentage Error. Key findings reveal the superiority of Hybrid ARIMA-ANN model than in other Models, for forecasting of Indian Arabica Plantation coffee price.