Abstract:This research paper entitled “Forecasting of Pigeon pea yield in Bihar through Auto-Regressive Integrated Moving Average Model (ARIMA)” is based on the secondary data. Data was collected for the years 1980 to 2021 from the websites and Department of Economics and Statistics and India Agri. Stat. The different ARIMA models are judged on the basis of Autocorrelation Function (ACF) and Partial autocorrelation Function (PACF) at various lags. The ARIMA models are fitted to the original time series data as well as the first difference data to check the stationarity. The possible ARIMA model are identified on the basis of significant coefficient of autoregressive and moving average components. Data for the year 1980 to 2019 were used for model building and rest two years data i.e. 2020 and 2021 were used for validation of yield forecast model of Pigeon pea.
It was found that the ARIMA (1, 1, 1) model is best fitted for forecasting of Pigeon pea yield in Bihar among all the other models namely ARIMA (0, 0, 1), ARIMA (1,0,1), ARIMA (0, 1, 1), ARIMA (1, 1, 1), ARIMA (0,1, 2), ARIMA (2, 0, 0) and ARIMA (2, 0, 1). The parameters of all these models were computed and tested for their significance. Various statistics were also computed for selecting the adequacy, invertibility, stationarity, Diagnostic checks, % forecast error and parsimonious model with t-test and chi-square test. The low values of MAPE, MAE, RMSE, % forecast error and BIC are good inferences for forecasting of Pigeon pea yield in Bihar. Five years ahead forecasting from 2020 to 2024 for pigeon pea yield were calculated using ARIMA models. The five year ahead yield forecasting of pigeon pea in Bihar for the year 2020, 2021, 2022, 2023 and 2024 are 1723 kg/ha, 1680 kg/ha, 1671kg/ha,1693 and 1707 kg/ha respectively with forecast error for the year 2020 and 2021 are 2.23% and 4.62% respectively.
The ARIMA model for forecasting yield of Pigeon pea in Bihar is as below:
Zt – Zt-1 = 16.019+ 0.485 (zt-1 - zt-2) –0.995 (at-1 - at-2) + at (for Bihar)