NARX based forecasting of petrol prices

The ongoing hike in the price of crude oil and its products has raised the concern to predict future prices of the petrol. This prediction is also useful in order to make the economy of a nation to be stable. Accuracy in the prediction of petrol price is the most important aspect. Our work here consists of theories on how dynamic neural networks along with auto regression can be applied for petrol price forecasting. The accuracy for calculation of petrol price is very crucial for making balancing in consumer producer demand. Profit, long and short term planning of production is depending on the accuracy of price forecasting and consumer can efficiently maximize their utilities. In this paper proposed algorithm is based on Non Linear Autoregressive model with exogenous input(NARX)for petrol price forecasting. Results are compare with minimum mean square error with different training algorithm.