Forecasting volatility of quality for wheat soya blend products in the statistical model of ARCH

Abstract

Present work has explored the impact of type of food products on testing for ARCH effects and on the estimation of ARCH models for food products analysis data. Our sample comprises physiochemical and microbial analysis data for food products. In our analysis the different value for different variables of parameters of the ARCH-LM test; the lags are 1. The corresponding p-value is >0.05, which is very high. So we have no difficulty to accept the null hypothesis of no ARCH error in the analysis series. The parameters of Wheat Soya Blend (WSB) analysis are insignificant that means no ARCH effects of the models. The estimation results are given in the table 3 shows that the values of Dickey–Fuller (DF) test for all variables p-value<0.05 at 5%, level of significance except Sugar (as sucrose) (%) and Standard Plate Count (cfu/g) which implies that the variables series is stationary. An outcome of DF test confirms that the physiochemical analysis variables series is stationary. Our results revealed that the ARCH model satisfactorily explains volatility and is the most appropriate model for explaining volatility in the series under analysis.