Optimal multi-variable grey forecast

Taking advantage of the characteristics of few data and poor information, grey system theory sets up differential equation model for accumulated generation series to forecast, which has been extensively used in many areas. In the forecast process of grey model, data sample size and variable number can affect forecast results. This paper puts forward a new method of optimal forecast variable number and data sample size for multi-variable grey model. The goal function is the minimum fitting relative error, and there are two constraints: one is data sample constraint; the other is variable number constraint. The algorithm can solve factor choice and data sample size determination problem, and fully use sample information. Case studies show that the method can produce good forecast results.