Sales forecasting based on support vector machines regression

Sales forecasting is one of the most crucial issues of enterprise operation. Sales forecasts, which form the basis for the planning of inventory levels, production scheduling and etc., are probably the biggest challenge in the manufacturing enterprises, which leads it a widely researched area. Generally, time series forecasting methods such as exponential smoothing are commonly used for sales forecasting and have made great contribution in practice, but efforts on looking for superior forecasting method are still made by practitioners and academia. This paper deals with the application of a novel neural network technique, support vector machines (SVMs), in sales forecasting. The objective of this paper is to examine the feasibility of SVMs in sales forecasting. A data set from a manufacturing firm in China is used for the experiment to test the validity of SVMs. The experiment shows SVMs a better method in sales forecasting.