
Research Article
Vietnamese Export Growth Prediction Applying MIDAS and MF-VAR on Mixed-Frequency Data
@INPROCEEDINGS{10.1007/978-3-030-92942-8_1, author={Nguyen Thi Hien and Hoang Anh Tuan and Dinh Thi Ha and Le Mai Trang and Tran Kim Anh and Dao The Son}, title={Vietnamese Export Growth Prediction Applying MIDAS and MF-VAR on Mixed-Frequency Data}, proceedings={Nature of Computation and Communication. 7th EAI International Conference, ICTCC 2021, Virtual Event, October 28--29, 2021, Proceedings}, proceedings_a={ICTCC}, year={2022}, month={1}, keywords={Export growth Forecasting Mixed-frequency data MF-VAR MIDAS}, doi={10.1007/978-3-030-92942-8_1} }
- Nguyen Thi Hien
Hoang Anh Tuan
Dinh Thi Ha
Le Mai Trang
Tran Kim Anh
Dao The Son
Year: 2022
Vietnamese Export Growth Prediction Applying MIDAS and MF-VAR on Mixed-Frequency Data
ICTCC
Springer
DOI: 10.1007/978-3-030-92942-8_1
Abstract
Import and export growth forecasting is always a concern for researchers as well as policymakers in any country. However, forecasting this index through methods that employ data sets with the same frequency does not reflect reality due to the fact that the economic and financial indicators have different published frequencies. Therefore, between two data periods, some important information affecting the target variable maybe not be included in the model, leading to the limited accuracy of the forecast. The approach of data analysis with multiple frequencies has gained a lot of interest recently in order to overcome data restrictions and increase forecasting performance. In this article, we study and apply a number of models for mixed-frequency data such as MF-VAR and MIDAS to predict Vietnam’s exports based on collected data sets in the period from 2006 to 2020. The findings show that the MF-VAR model has high forecasting error and is not suitable for predicting the export growth of Vietnam, while the MIDAS model gives good prediction results on the same data set. In addition, the prediction results also reveal that the MIDAS model is effective for short-term forecasting. This result is quite similar to that of several previously published studies. The findings of this research also open a promising direction in studying and applying the mixed frequency data models to forecast export growth as well as other economic indicators .