TIME SERIES DATA PREDICTION OF NATURAL GAS CONSUMPTION USING ARIMA MODEL

Time series data modeling and prediction are used in various practical domains. Thus a lot of active research works is going on in this subject during several years. Several important models have been proposed in literature for accuracy and efficiency of time series modeling. The aim of this article is to present a statistical time series prediction using Autoregressive Integrated Moving Average (ARIMA) models. We have explained here different statistical methods of time series models. Here we have collected historical data of Natural Gas Consumption in India from year 2005 to 2014 of every quarter's data. Fitting a model to a dataset is used Goodness of fit statistic. Here the model used preliminary estimation: Yule-Walker prediction accuracy models fitted to a time series. We have shown the obtained prediction diagram, which graphically describes prediction of Natural Gas consumption in India. In this article, our observation about different methods of time series modeling and prediction are explained clearly. We have also shown that components such as trends and periodicity in the time series can be explicitly modeled, with the data being decomposed into trend, seasonal and residual components.