TY - JOUR
T1 - Time Variation in Diversification Benefits of Commodity, REITs, and TIPS
AU - Huang, Jing zhi
AU - Zhong, Zhaodong Ken
N1 - Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2013/1
Y1 - 2013/1
N2 - Diversification benefits of three "hot" asset classes-Commodity, Real Estate Investment Trusts (REITs), and Treasury Inflation-Protected Securities (TIPS)-are well-studied on an individual basis and in a static setting. Using data from 1970 to 2010, this paper documents both that the three asset classes are in general not substitutes for each other, and that diversification benefits of each asset class change substantially over time. Therefore, all three asset classes ought to be included in investors' portfolios. Furthermore, we show that the observed time variation in diversification benefits can be explained by time-varying return correlations. To see the implications of these findings for asset allocation in practice, we examine the out-of-sample performance of portfolio strategies, based on a variety of correlation structures. We find that the Dynamic Conditional Correlation (DCC) model (Engle, J Bus Econ Stat 20(3):339-350, 2002) outperforms other correlation structures, such as rolling-window, historical, and constant correlations. Our findings suggest that diversification benefits of the three asset classes should be examined in a dynamic setting, and that investors need to use appropriate correlation estimates to adjust for such time variation.
AB - Diversification benefits of three "hot" asset classes-Commodity, Real Estate Investment Trusts (REITs), and Treasury Inflation-Protected Securities (TIPS)-are well-studied on an individual basis and in a static setting. Using data from 1970 to 2010, this paper documents both that the three asset classes are in general not substitutes for each other, and that diversification benefits of each asset class change substantially over time. Therefore, all three asset classes ought to be included in investors' portfolios. Furthermore, we show that the observed time variation in diversification benefits can be explained by time-varying return correlations. To see the implications of these findings for asset allocation in practice, we examine the out-of-sample performance of portfolio strategies, based on a variety of correlation structures. We find that the Dynamic Conditional Correlation (DCC) model (Engle, J Bus Econ Stat 20(3):339-350, 2002) outperforms other correlation structures, such as rolling-window, historical, and constant correlations. Our findings suggest that diversification benefits of the three asset classes should be examined in a dynamic setting, and that investors need to use appropriate correlation estimates to adjust for such time variation.
UR - http://www.scopus.com/inward/record.url?scp=84871946593&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84871946593&partnerID=8YFLogxK
U2 - 10.1007/s11146-011-9311-6
DO - 10.1007/s11146-011-9311-6
M3 - Article
AN - SCOPUS:84871946593
SN - 0895-5638
VL - 46
SP - 152
EP - 192
JO - Journal of Real Estate Finance and Economics
JF - Journal of Real Estate Finance and Economics
IS - 1
ER -