A comprehensive analysis of price changes in crude oil data involving time series modelling

This research model explored different time series modelling approach over Crude oil prices. In time series analysis, we assume that the current price of crude oil reflects the effect of all the influencing factors. So that the price forecasting of the crude oil can be done using the past crude oil prices. The main assumption in this time series modeling is that the past crude oil prices can be used to predict the future crude oil price. Although the time series analysis can find the trend, there will be limitations to the forecasting capability of the model that we use in the analysis when the reversal in trend is observed in the data taken or the pattern repeated may not be followed by the future prices. Different types of trend patterns such as increasing trend, decreasing trend or periodic patterns can be obtained. Time series analysis is more useful and will give better forecasting only when the data follows any of these trends. In this work, data analysis on crude oil data set is performed. A novel time-series forecasting approach based on Auto-Regressive Integrated Moving Average (ARIMA) model, Seasonal Auto-Regressive Moving Average (SARIMA), ARCH (Auto Regressive Conditional Heteroscedasticity) model, GARCH (Generalized Auto Regressive Conditional Heteroscedasticity) are also proposed using the R programming language for statistical computing and graphics. The results will help the researchers from various community to gauge the trend and improvise containment strategies accordingly.