Transport infrastructure equities in mixed-asset portfolios: estimating risk with a Garch-Copula CVaR model. Issue 2 (3rd April 2018)

• Record Type: Journal Article
• Authors: Chakkalakal, Louis; Hommel, Ulrich; Li, Wenwei
• Is Part Of: Journal of property research. Volume 35:Issue 2(2018)
• Page Start: 117
• Page End: 138
• Publication Date: 2018-04-03
• Subjects: Transport infrastructure -- portfolio optimisation -- risk management -- VaR/CVaR -- GARCH-Copula; Real estate business -- Great Britain -- Periodicals; Land use -- Great Britain -- Periodicals; 333.3
• DOI: 10.1080/09599916.2018.1461126
• Languages: English
• ISSNs: 0959-9916

Abstract

Abstract: Transport infrastructure is an important subsector within infrastructure, but knowledge of its equities in terms of risk-return characteristics and contribution to portfolio performance is still limited. This study assesses the subsector individually and in a multi-asset, index-based portfolio. In doing so, we apply a t-Copula-based Conditional Value-at-Risk model to simulate risk and returns. Our findings reveal that the subsector has a relatively low dependency on other equities, performs like other alternative asset classes such as general real estate, and does not grant significant risk diversification benefits for mainstream institutional investors such as pension funds. Investors aiming for higher target returns may however assign substantial weights to transport infrastructure, supporting our conjecture that it does not share the same asset class characteristics as general infrastructure. By contrasting Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR) scores for both the mean-variance framework and the t-Copula simulation, we further document the limitations of traditional VaR approaches. Hence, this study's results support the use of risk assessment tools that incorporate non-normal distributions to represent multivariate dependence structures.