Forecasting of any issues, events or variables requires an in-depth understanding of the underlying factors affecting it. Such is the case for forecasting annual productivity of pulse crops. India's ubiquitous position as the leading producer, the foremost consumer and the largest importer of pulses is besmirched by abysmally mediocre policy intervention and equally unimpressive agricultural research budgets. Pulses in India recorded less than 40 per cent growth in production in the past 40

more »

... s while its per capita availability declined from 60 grams a day in the 1950 to 39.4 grams a day in 2011. Pulses productivity, in the context of India, extensively depends upon numerous factors namely: good rainfall, timely use of appropriate fertilizer and pesticides, favourable climate and environments etc. Currently, even as production has stabilized at 18.5 million tones, our consumption is hovering at 22 million tones, which necessitates yearly pulse imports of around 3.5-4 million tones. Therefore, forecasting productivity of pulse crop is indispensible, as large chunk of people depends on agriculture for their livelihood. Various uni-variate and multi-variate time series techniques can be applied for forecasting such variables. In this paper, ARIMA model has been applied to forecast annual productivity of selected pulse crops. For empirical analysis a set of different has been considered, contingent upon availability of required data. Applying annual data from 1950-51 to 2014-15, forecasted values has been obtained for another 5 years since 2016. The evaluation of forecasting of pulses production has been carried out with root mean squares prediction error (RMSPE), mean absolute prediction error (MAPE) and relative mean absolute prediction error (RMAPE). The residuals of the fitted models were used for the diagnostic checking. These forecasts would be helpful for the policy makers to foresee ahead of time the future requirements of grain storage, import and/or export and adopt appropriate measures in this regard.