A Dynamic Analysis of the Bank of Japan’s ETF/REIT Purchase Program (Forthcoming in “Expert Systems With Applications”)

Abstract

This paper provides a time series analysis of the Bank of Japan (BOJ)’s exchange traded fund (ETF) and real estate investment trust (REIT) purchase program. The program is a part of the BOJ’s unconventional monetary policy introduced in December 2010 and the timing and quantity of purchases have changed substantially over time. To the best of our knowledge, the current study is the first attempt at investigating the purchase program from a dynamic viewpoint and developing a system that takes expert knowledge as input to analyze the BOJ’s purchase program and outputs a trading strategy.
Firstly, to find factors underlying decisions of the BOJ’s purchases, we apply a state space model with explanatory variables obtained by expert knowledge, which is an extension of a logistic regression and enables us to predict their timings with high accuracy throughout all periods. Secondly, we figure out the dynamics of the program’s effect on equity and REIT prices by using a different type of state space model. Finally, we find a profitable trading strategy that incorporates the insights revealed in the above dynamic analysis and that focuses on an effect of the BOJ’s purchase on the spread between the Tokyo Stock Price Index (TOPIX) and Nikkei225 Index (Nikkei225). In addition, we implement event studies, particularly the price dynamics of ETF and REIT just before and after the purchase.

The preprint version is available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4053708