A short-term macroeconomic Forecasting model based on GMDH

Abstract:

Traditional shortterm macroeconomic forecasting research is usually based on quantitative or qualitative forecasting models, while GMDH algorithm has both qualitative and quantitative characteristics. This work proposes a mixed mathematical prediction model to forecast shortterm macroeconomic. The model is a combination of GMDH autoregressive model and quadratic autoregressive model. Relevant statistical data are used for empirical analysis, and the results are compared with the traditional GMDH model and the quadratic autoregressive model. The results show that the hybrid forecasting model proposed in this paper has good forecasting accuracy and is an effective means of economic forecasting.