Sales forecasting of computer products based on variable selection scheme and support vector regression

摘要: Since computer products are highly replaceable and consumer demand often changes dramatically with the invention of new products, sales forecasting is therefore always crucial for product management. When constructing a model, discussing understanding important predictor variables can help focus on improving management efficacy. Aiming at to select appropriate variable construct effective this study combines selection method support vector regression (SVR) hybrid model products. In order evaluate feasibility performance proposed approach, compiles weekly data five including Notebook (NB), Liquid Crystal Display (LCD), Main Board (MB), Hard Disk (HD), Card (DC) from retailer as illustrative example. The experimental results indicate that scheme not only provide better result than four competing models in terms error, but also exhibit capability identifying variables. Furthermore, useful information be provided by identified different thereby increasing