Annual Automobile Sales Prediction Using ARIMA Model

Abstract

Sales forecasting is a most important application in industries and has been one of the most scientifically and technologically challenging problems around the world. One approach of prediction is to spot patterns in the past, when it is known in advance what followed them and verify it on more recent data. If a pattern is followed by the same outcome frequently enough, it can be concluded that it is a genuine relationship. Because this approach does not assume any special knowledge or form of the regularities, the method is quite general applicable to other series not just climate. Sales prediction phenomena have many parameters like Number of sales, production, Consumed cost and Time required that are impossible to enumerate and measure. In this paper, we are going to use the ARIMA model for predicting the number of sales for a Time series data. The dataset tractor sales data for a period of ten years (2003-2014) obtained from the Mahindra Tractors Company are used from which use to classify the performance by drawing various scattered plots and graphs. The result of the ARIMA results shows that which predicts better for the sales prediction of the next following 5 years.