Using BDTI’s High-Dimensional CG Big Data to Predict TSR in Japan (working paper)

S&P-topix - BDTI director training

This is the first draft of a working paper led by two respected Japanese academicians who used governance and firm-specific big data to predict equity returns in Japan, using machine-learning techniques. Conclusion: “we constructed a prediction model of firms’ future TSR and used it to show that the investment strategy based on the model’s predictions could generate non-negligible improvement in returns. These results suggest that high-dimensional corporate governance variables contain informative signals associated with future firm performance over and above reliance on purely financial data.”

The research was conducted using BDTI’s detailed, Japan-specific time-series database for all listed companies in Japan. The results are consistent with the fact that every fund manager that has backtested our data so far has bought a license, and every licensee has renewed so far. It would seem that Japan’s square peg of three different governance structures and peculiar practices does not seem to fit into the standard “global” round hole framework used by other data providers.

This paper has been accepted for presentation at the Japanese Economic Association’s Spring Conference for 2023. However, there may be changes to this draft.

We are now planning to hold a webinar where the Professors will present their results and we will discuss the implications.  Most likely this webinar will be held in the late afternoon of Tuesday June 13th. Please send an email to if you may be interested to view this webinar.

This research, and the paper, was prepared as a project within the scope of BDTI “public interest activity” to conduct research on corporate governance.  We are thankful to our Donors and Supporters for making this research possible.

Nicholas Benes
CEO, The Board Director Training Institute of Japan

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.