Forecasting Exports of Tea from India: Application of Arima Model

Abstract

Exports form an integral part of decision making for a large country like India. Tea is an important beverage which is homogeneously essential in the domestic sector and across the globe, transcending geographical boundaries, social groups and ages. About seventy three percent of the world's tea exports come from four countries, including India. This paper uses time series model namely ARIMA to make short term forecasting of tea exports in India using 79 monthly observations. The battery of diagnostic tests are conducted to examine the efficacy of the model hence built. The model ARIMA (1,1,0) has the lowest AIC and BIC criteria, since it has two parameters following the principle of parsimony ,this model is chosen An eight period ahead export of tea is predicted. The observations indicate a rising trend in exports. The model fitting is compared with SES and HES to show that ARIMA has higher fitting accuracy than exponential smoothing. The implications of the results developed in the paper is useful for considering short term market fluctuations in export of tea in India.
JEL Classification : C 53, F 17, F 23.