Research On China-ASEAN Import And Export Trade Volume Based On Multiple Forecast Model

Import and export trade is an important tool for stimulating national economic growth and achieving foreign exchanges.ASEAN is the third largest partner of China's foreign trade.Analyzing and forecasting China's import and export volume with ASEAN is also a part of the economic integration of China-ASEAN marine big data platform.At present,more methods based on econometrics are used,but with the continuous development and maturity of machine learning and data mining technologies,a lot of classical algorithms have been settled in practice.Therefore,this paper combines machine learning methods to explore a forecasting model that can effectively forecast the volume of China-ASEAN import and export trade.It not only realizes the demand of the China-ASEAN marine big data platform for foreign trade analysis,but also provides intellectual support for China's macroeconomic regulation of ASEAN's economic and trade cooperation.This article first analyzes the economic indicators that affect China-ASEAN import and export trade volume,and uses principal component analysis methods to analyze the impact factors,and obtain seven impact factors that can replace the original economic indicators with over 99.9%of information volume.Subsequently,a prediction model of China-ASEAN import and export trade volume based on linear regression was proposed.Based on multiple linear regression,ridge-based regression,and elastic network regression model,the historical data of China-ASEAN import and export trade volume and influence factors were trained to build prediction models.Calculate the future import and export volume of China and ASEAN through the estimated values of the seven economic factors.Then,as China-ASEAN import and export trade is the result of a combination of factors such as economic indicators,bilateral relations,and policy influences,the historical data of China-ASEAN import and export trade volume is converted into a time series,and the internal data change rules are tapped.Time series ARIMA prediction model.Secondly,for digging deeply into the data sequence changes of China-ASEAN import and export trade,we propose a prediction model based on LSTM network that can learn long-term information-dependent capabilities.Thirdly,for forecasting China-ASEAN import and export trade volume more accurately,a combination forecast model based on time series ARIMA and LSTM network is proposed based on the idea of data preprocessing and optimization parameter combination forecasting.Experiments have shown that it predicts that the accuracy and correlation coefficient of China-ASEAN import and export trade volume have been significantly improved.Therefore,based on this forecasting model,the monthly import and export trade volume between China and ASEAN in the short term is predicted.Finally,the research results of this paper are applied to the module of China-ASEAN marine big data platform economic comprehensive database.This module can help platform users understand the historical data of import and export trade between China and ASEAN countries.It can allow users to predict the annual import and export trade volume of China and ASEAN through the annual projections of various economic indicators,or it can directly Monthly forecast of import and export trade with ASEAN.Through practical application,the effectiveness of the predictive model for China-ASEAN import and export trade was tested.