Macroeconomic Short-Term High-Precision Combined Forecasting Algorithm Based on Grey Model

Cao Li,Jian Su

doi:10.1155/2021/7026064

Abstract

Using the characteristics of grey forecasting, which requires a small amount of sample data and a simple modeling process, to predict the main macroeconomic indicators in the early stage, combined with the filtering decomposition method and the production function method, establishes a short-term high-precision combination forecasting algorithm for macroeconomics based on the grey model. The algorithm uses the improved HP filter method in the HP filter method to study whether the potential economic growth rate can be more accurately measured, and the production function method is used to calculate the potential economic growth rate. First, the two methods are used to calculate the potential economic growth rate. The accuracy of this method finally established a combined model based on the two models for short-term forecasting. Under the premise of considering economic factors, the input data is preprocessed, and the high-precision combined forecast is used to finally obtain the macroeconomic forecast results. The calculation examples in the paper show that the method is feasible and effective.

Highlights

Accurate forecasting of the macroeconomy is a necessary prerequisite for the country’s correct regulation and control of the macroeconomy
Macroeconomic indexes include gross domestic product (GDP), consumer price index (CPI), and retail price index (RPI). e gross domestic product (GDP) is the total value of the final products and services produced by a country or region in a certain period of time
E index can understand the economic situation of a country or region from a macro perspective. e consumer price index (CPI) reflects the price situation at the consumer level, and the index can reflect the degree of inflation or deflation to a certain extent. e retail price index of commodities (RPI) reflects the price changes of retail commodities, which can be used to understand the supply and demand of retail commodities in the market

Summary

Introduction

Accurate forecasting of the macroeconomy is a necessary prerequisite for the country’s correct regulation and control of the macroeconomy. Grey system theory can still deal with many uncertain problems very well when the amount of information and data is relatively small. Erefore, the grey system theory is usually considered as an advantageous theoretical method that can be used to solve some uncertain problems with characteristics such as a small amount of information and incomplete and discrete data. Zhang et al used the mixed frequency Bayes vector autoregressive model (MFBVAR) to predict China’s macroeconomic indicators. Is paper combines the GM model with the combination algorithm and uses the GM (1, 1) model with different trade-offs on historical data to make preliminary predictions, and these prediction results and the relevant economic factor parameters for each year are used as the high-precision combination prediction algorithm entering the final prediction result Macroeconomic forecasting is a typical nonlinear problem. e reason is that the macroeconomic operation is affected by many factors, such as changes in policies and the level of economic development. erefore, these factors must be considered in the process of forecasting. is paper combines the GM model with the combination algorithm and uses the GM (1, 1) model with different trade-offs on historical data to make preliminary predictions, and these prediction results and the relevant economic factor parameters for each year are used as the high-precision combination prediction algorithm entering the final prediction result

Grey Prediction Model

Macroeconomic Short-Term High-Precision Combination Forecasting Algorithm

China’s Potential Economic Growth Rate Calculation

Findings

Conclusion