Forecasting Coal Production in India: A Time Series Approach
Avni Gangwar, Diksha Rathor, Praveen Kumar TrıpathıThis article is intended to produce the forecasts for coal production in India through some time series models. This study describes the component-based and correlation-based time series models for its purpose. The separate analyses were performed by applying Naïve, Holt’s and ARIMA models on a real data set based on the coal production in India between 1980 and 2022. On the basis of the retrospective predictions and accuracy measure results, an ARIMA (2,2,2) model was selected as a good choice for the data in hand. A particular ARIMA (2,2,2) model was selected by using the AIC and BIC of model selection. For the validity of the finally selected ARIMA (2,2,2) model, a residual diagnostics check has been performed; and the future predictions have been made for the next 5 years. Such an analysis is expected to add some new approaches in the literature of forecasting the energy sources, especially with reference to India.
