Research Article
Forecasting Foreign Trade Trends Based on Combined ARIMA and Composite Quantile Regression Models
@INPROCEEDINGS{10.4108/eai.26-5-2023.2334441, author={Luwei Zhang}, title={Forecasting Foreign Trade Trends Based on Combined ARIMA and Composite Quantile Regression Models}, proceedings={Proceedings of the 2nd International Conference on Mathematical Statistics and Economic Analysis, MSEA 2023, May 26--28, 2023, Nanjing, China}, publisher={EAI}, proceedings_a={MSEA}, year={2023}, month={7}, keywords={arima model composite quantile regression foreign trade trend forecasting}, doi={10.4108/eai.26-5-2023.2334441} }
- Luwei Zhang
Year: 2023
Forecasting Foreign Trade Trends Based on Combined ARIMA and Composite Quantile Regression Models
MSEA
EAI
DOI: 10.4108/eai.26-5-2023.2334441
Abstract
Predicting the trend of total foreign trade is vital for studying the national economic system. In this paper, after using a full subset regression model for variable screening, a combined model of ARIMA and composite quantile regression (CQR) estimation series is used to predict the trend of total foreign trade. This model combines internal and external factors to make forecasts, which has higher accuracy and stronger stability over a short period at specific accuracy values, and can obtain better forecasting results than a single model. In practice, the combined model used in this paper can give full play to the influence of specific factors, thus providing a more accurate and robust forecasting model for scientific and rational decision-making.