CEU Electronic Theses and Dissertations, 2020
| Author | Nguyen, Mai Hao |
|---|---|
| Title | Using machine learning to estimate impacts of free trade agreements on export performance: A case of East Asian economies |
| Summary | This study uses two machine learning methods, namely text as data analysis and principal component analysis to create two indicators of trade facilitation depth of legal commitments, namely Wordscores estimates and principal component analysis depth (PCA depth), and compare these two indicators with simple additive scores. Each of the three indicators, together with bilateral tariff, serves a predictor in a gravity equation in order to study the impacts of qualitative and quantitative commitments on the export of seventeen East Asian economies. The results suggest that Wordscores estimates and PCA depth, instead of additive scores, are a better representation of trade facilitation depth. While tariff maintains a significant negative correlation with export, trade facilitation depth of legal commitments shows an insignificant positive relationship with export for the full sample of East Asian economies. However, the correlation between trade facilitation depth and export becomes significant and positive for a sub-sample of six East Asian exporters with high implementation rates of trade facilitation measures, but still insignificant and negative for the rest of East Asian exporters. |
| Supervisor | Brown, Caitlin Karagiannis, Yannis |
| Department | Public Policy MA |
| Full text | https://www.etd.ceu.edu/2020/nguyen_mai-hao.pdf |
Visit the CEU Library.
© 2007-2021, Central European University