Daily Capital Control Index: Powered by Machine Learning
Joint work with: Prof. Roberto Samaniego This project focuses on the development of a high-frequency Daily Capital Control Index for 119 countries, spanning from January 1, 2000, to the present. The index tracks six categories of capital account interventions, offering a real-time tool for the tracking and analysis of global capital control policies. By providing timely and granular insights, the index serves as a critical resource for researchers, policymakers, and market participants seeking to understand cross-country differences in capital control measures. To ensure precision, the index employs machine learning techniques, including Linear Regression and LASSO, trained on the Ka-open Index from the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER). The project also includes the development of a comprehensive dataset and a user-friendly website that enables real-time updates and dynamic access to the dataset, facilitating seamless access to up-to-date capital control information.