M S Khan
Kalinga Institute of Industrial Technology, India
Title: Prediction of petroleum price in India
Biography
Biography: M S Khan
Abstract
Forecasting of oil prices has always been a matter of great importance due to its influence in driving a country’s economy. As a matter of fact, the petroleum industry is considered to be the biggest contributor in the industrial sector in terms of providing raw materials to the other industries and generating revenues. Due to the non-linear and unpredictable nature of the oil prices, a lot of forecasting techniques have been developed and used to check whether they are capable of forecasting the oil prices satisfactorily. In this paper, two widely used techniques, ARIMA analysis and GMDH neural network has been used to forecast the prices of four petroleum products such as petrol, diesel, LPG and kerosene in India for a three month period (February 2015 to April 2015). The results obtained are compared with the actual prices for the above time period. It is observed that the overall accuracy considering all the four petroleum products shows promising results thus justifying them capable of forecasting the prices of the different petroleum products in India. When both the techniques are compared, ARIMA modelling shows better results (97.99% accuracy) as compared to that of the GMDH neural network method which is 96.97%.