DETERMINING FDI INFLOWS IN INDIA: USING BOX-JENKINS ARIMA APPROACH

Using time series data for FDI inflow in India from 1991 to 2021, the study seeks to model and predict the FDI inflows in India. The Autoregressive integrated moving average (ARIMA) model created by Box and Jenkins (1976) was utilised to develop the model. Identification of the UBJ included determining the appropriate AR (autoregressive) and MA (moving-average) polynomial orders, i.e., p and q values. The rankings were used to determine the stationary series' autocorrelation and partial autocorrelation functions. It was determined that FDI data were not static and that a single-order difference was sufficient to create the required stationary series. The study identified a low BIC value and then proposed the ARIMA model (0,1,2) as an appropriate FDI predictor in India. The expected FDI inflows for 2022–23 through 2029-2030 were within the confidence interval. The percentage variation between predicted and observed numbers assures that our forecast prices are near actual prices.