Author(s): LK Kumawat, PB Marviya, S Kumawat, GK Chaudhary, DM Padalia, Tejas S and IB Savaliya

Abstract: The present study was carried out to estimate the trends of area, production and productivity of cumin in Banaskantha district of Gujarat. The time series data on area, production and productivity of cumin of Banaskantha district for the period of 2000-01 to 2022-23 were collected from the published reports by Directorate of Agriculture, Gujarat state, Gandhinagar. The data from 2000-01 to 2018-19 for cumin were used for model building and remaining data 2019-20 to 2022-23 for validation of the forecast model. An attempt was made in present investigation to fit polynomial and ARIMA models to arrive at a methodology that can precisely explain the fluctuation in area, production and productivity of cumin in Banaskantha district of Gujarat and to compare different models. The first, second and third degree polynomial models were fitted on original data as well as three, four and five year moving average data approach for the area, production and productivity. The Autoregressive Integrated Moving Average (ARIMA) models were fitted to original time series data after checking the stationary condition of the data. Among the fitted polynomial models, the suitable model was identified on the basis of significance of regression coefficient, adjusted R2, root mean square error (RMSE); mean absolute error (MAE), normality by the Shapiro-Wilk test and randomness of residuals by the Run test. Different orders of ARIMA models (p, d, q) were judged on the basis of autocorrelation function (ACF) and partial autocorrelation function (PACF) at various lags. Different possible ARIMA models were fitted and from these, the models were selected on the basis of significant autoregressive and moving average term, Akaike’s Information Criteria (AIC), Schwartz-Bayesian Criteria (SBC) values, adjusted R2, normality of residuals by Shapiro-Wilk (1965) test and Box-Ljung (1978) test. For the trend, cubic model with five year moving average data approach was best suitable polynomial model for area and production of cumin in Banaskantha district of Gujarat. For the trend, quadratic model with five year moving average data approach was best suitable polynomial models for productivity of cumin in Banaskantha district of Gujarat. Among the ARIMA models, ARIMA (2, 1, 2) model was found suitable to explain the pattern of cumin productivity. None of the ARIMA models were suitable for area and production of cumin due to lack of one or more criteria for selection of models. Thus, in general because of crucial requirements of model selection criteria in polynomial as well as ARIMA models, few models could get selected.

DOI: 10.22271/maths.2024.v9.i5Sb.1810

Pages: 67-80 | Views: 179 | Downloads: 16

Download Full Article: Click Here