The dynamic role of the Japanese property sector REITs in mixed-assets portfolio

Abstract

PurposeThis study examined the dynamic role of the Japanese property sector, particularly the real estate investment trusts (REITs), in mixed-asset portfolios of stocks and bonds, as well as office, retail, hotel and residential REITs.Design/methodology/approachDaily data were retrieved from 01 January 2008 to 31 December 2019. The sample time frame consisted of in-sample and out-of-sample periods. The dynamic conditional correlation-generalised autoregressive conditional heteroskedastic (DCC-GJRGARCH) model was deployed to obtain the forecast estimates of time-varying volatility of REITs and correlations with other assets. The estimates were employed to construct out-of-sample portfolios based on the three assets for daily investment. The five sets of portfolios with each individual property sector REITs, as well as a portfolio of stocks and bonds that served as a benchmark, were produced. The average utility for each set of portfolios was estimated and compared with the average utility of the benchmark portfolio. The average transaction cost (TC) for portfolio rebalancing was calculated as well.FindingsThe forecast of volatility estimates for each property sector revealed that each asset displayed a similar pattern with the differences in the volatility magnitude. Notably, hotel and retail REITs were more volatile than other property sector REITs. The property sector REITs exhibited a positive correlation with stocks but negatively linked with bonds. The results unveiled the diversification benefits of incorporating property sector REITs. The portfolio with property sector REITs had higher risk-adjusted returns and utility, compared to portfolio consisting of stocks and bonds. The benefits outweighed the TC for portfolio rebalancing.Practical implicationsThis study highlights the importance of quantifying the conditional time-varying volatility and correlations of the property sector REITs with other asset returns, especially for investment decision, to select and include property sector REITs in mixed-asset portfolios. For fund managers seeking liquid assets in daily investment, this analysis suggests the inclusion of hotel and retail REITs to enhance REITs' portfolio performance.Originality/valueThis study is the first to investigate the dynamic characteristics of the volatility and correlation of each property sector REITs with other financial assets by employing the conditional framework that accounted for short- and long-run persistency in economic shocks. The reported outcomes shed light on the differences in the underlying properties that contribute to the variances in dynamic volatility of each sector REITs, as well as REITs' correlations with stocks and bonds. This application enables the authors to transmit the dynamics of variance-covariance matrix amongst each property sector REITs, stocks and bonds into asset allocation problem on a daily basis.

Analyzing Portfolio Optimization in Cryptocurrency Markets: A Comparative Study of Short-Term Investment Strategies Using High-Frequency Data

“…Secondly, the correlation matrix filters out random and systemic co-movements in constructing portfolios, improving risk profiles (Zema et al, 2021). It enables capturing conditional volatility and correlation for forecasting future variance-covariance matrices between assets for optimal portfolio construction (Razak, 2023). The correlation matrix is a fundamental component in portfolio optimization models, and its accurate estimation is crucial for their effectiveness (Tola et al, 2008).…”