The grey prediction model is a prediction method that establishes a mathematical model to make predictions through a small amount of incomplete information. It is based on the past and present development laws of objective things, with the help of scientific methods to describe and analyze the future development trends and conditions, and to form scientific assumptions and judgments.
This project mainly uses the grey forecasting model GM(1,1) to forecast GDP forecasts, real estate consumer price index forecasts, and of course other time series data. The purpose of writing this algorithm is to record and "introduce others".
When -a<=0.3, the one-step prediction accuracy of GM(1,1) is more than 98%, and the three-step and five-step prediction accuracy is more than 97%, which can be used for medium and long-term prediction. When 0.3<=-a<=0.5, the one-step and two-step prediction accuracy of GM(1,1) is more than 90%, and the ten-step prediction accuracy is more than 80%, which can be used for short-term prediction.