This paper examines the link between risk and institutional quality, an unresolved issue in finan... more This paper examines the link between risk and institutional quality, an unresolved issue in finance. Our hypothesis is that institutions affect risk through extreme events and less through volatility. We focus on relative tail risk with an original approach that is able to estimate historical tail risk with greater precision. Using international stock market data, we show that tail risk is stable over time, unlike volatility. We find that tail risk captures the relation between risk and institutional quality better than volatility. Better governa
This study introduces a new technique to analyse the evolution of correlations for multiple time ... more This study introduces a new technique to analyse the evolution of correlations for multiple time series. The technique is based on applying Topological Data Analysis (TDA) and we use it to gain insights about the evolution of commodity futures markets over the 1997-2017 period. Our findings complement the existing literature and provide new insights into the dynamics of commodity futures markets in the past two decades. Our analysis has both global and local aspects and could be applied to detect changes in correlation structure in a variety of time series as well as cross sectional settings.
. This paper provides a realization of K-theory with R/Z coefficients and proves an R/Z index th... more . This paper provides a realization of K-theory with R/Z coefficients and proves an R/Z index theorem.
We thank Keng-Yu Ho for research assistance. We would also like to thank Wayne Ferson for helpful... more We thank Keng-Yu Ho for research assistance. We would also like to thank Wayne Ferson for helpful discussions. We also thank Kenneth French and Martin Lettau for making available the data used in this study. All remaining errors are ours. 1 Portfolio-Based Tests of Conditional Asset Pricing Models
Abstract In this paper we focus on the optimal use of a variety of predictive variables to constr... more Abstract In this paper we focus on the optimal use of a variety of predictive variables to construct actively managed strategies using the components of the Dow Jones Industrial Average, and analyze whether these outperform the index itself. Our strategies are unconditionally e‐cient with the portfolio weights being pre-specifled functions of the predictive variables, and are thus both theoretically optimal and also relatively simple to implement. Our strategies signiflcantly outperform not only the DJIA but also the corresponding ‘flxed-weight’ strategies derived from classic mean-variance theory. For example, an optimal stock-picking strategy using term spread and convexity as predictive variables has an alpha of 11.8%, while the corresponding flxed-weight strategy barely matches the index. Similarly, an optimal market-timing strategy again using the same predictor variables achieves an alpha of 10.9%, whereas the corresponding flxed-weight strategy under-performs the index by 3.4...
ABSTRACT We provide a new approach to hedge fund replication that seeks to capture the non-linear... more ABSTRACT We provide a new approach to hedge fund replication that seeks to capture the non-linear dynamic factor exposures of hedge funds. Our approach constructs optimal benchmarks using liquid traded factors for a variety of hedge fund categories. We find that our benchmarks achieve good Sharpe ratios for a number of categories and are able to match the performance of about the 35 th percentile of individual funds before fees. Thus, our approach provides a low-cost alternative to capturing the dynamic non-linear factor exposures of hedge funds. We show that most of the gains from hedge funds before and during the credit-crunch crisis could have been matched or even beaten by simply exploiting the information contained in publicly observable economic indicators, calling into question the issue of hedge fund manager skills.
Journal of Financial and Quantitative Analysis, 2012
In this paper we study the economic value and statistical significance of asset return predictabi... more In this paper we study the economic value and statistical significance of asset return predictability, based on a wide range of commonly used predictive variables. We assess the performance of dynamic, unconditionally efficient strategies, first studied by Hansen and Richard (1987) and Ferson and Siegel (2001), using a test that has both an intuitive economic interpretation and known statistical properties. We find that using the lagged term spread, credit spread, and inflation significantly improves the risk-return trade-off. Our strategies consistently outperform efficient buy-and-hold strategies, both in and out of sample, and they also incur lower transactions costs than traditional conditionally efficient strategies.
This paper examines the link between risk and institutional quality, an unresolved issue in finan... more This paper examines the link between risk and institutional quality, an unresolved issue in finance. Our hypothesis is that institutions affect risk through extreme events and less through volatility. We focus on relative tail risk with an original approach that is able to estimate historical tail risk with greater precision. Using international stock market data, we show that tail risk is stable over time, unlike volatility. We find that tail risk captures the relation between risk and institutional quality better than volatility. Better governa
This study introduces a new technique to analyse the evolution of correlations for multiple time ... more This study introduces a new technique to analyse the evolution of correlations for multiple time series. The technique is based on applying Topological Data Analysis (TDA) and we use it to gain insights about the evolution of commodity futures markets over the 1997-2017 period. Our findings complement the existing literature and provide new insights into the dynamics of commodity futures markets in the past two decades. Our analysis has both global and local aspects and could be applied to detect changes in correlation structure in a variety of time series as well as cross sectional settings.
. This paper provides a realization of K-theory with R/Z coefficients and proves an R/Z index th... more . This paper provides a realization of K-theory with R/Z coefficients and proves an R/Z index theorem.
We thank Keng-Yu Ho for research assistance. We would also like to thank Wayne Ferson for helpful... more We thank Keng-Yu Ho for research assistance. We would also like to thank Wayne Ferson for helpful discussions. We also thank Kenneth French and Martin Lettau for making available the data used in this study. All remaining errors are ours. 1 Portfolio-Based Tests of Conditional Asset Pricing Models
Abstract In this paper we focus on the optimal use of a variety of predictive variables to constr... more Abstract In this paper we focus on the optimal use of a variety of predictive variables to construct actively managed strategies using the components of the Dow Jones Industrial Average, and analyze whether these outperform the index itself. Our strategies are unconditionally e‐cient with the portfolio weights being pre-specifled functions of the predictive variables, and are thus both theoretically optimal and also relatively simple to implement. Our strategies signiflcantly outperform not only the DJIA but also the corresponding ‘flxed-weight’ strategies derived from classic mean-variance theory. For example, an optimal stock-picking strategy using term spread and convexity as predictive variables has an alpha of 11.8%, while the corresponding flxed-weight strategy barely matches the index. Similarly, an optimal market-timing strategy again using the same predictor variables achieves an alpha of 10.9%, whereas the corresponding flxed-weight strategy under-performs the index by 3.4...
ABSTRACT We provide a new approach to hedge fund replication that seeks to capture the non-linear... more ABSTRACT We provide a new approach to hedge fund replication that seeks to capture the non-linear dynamic factor exposures of hedge funds. Our approach constructs optimal benchmarks using liquid traded factors for a variety of hedge fund categories. We find that our benchmarks achieve good Sharpe ratios for a number of categories and are able to match the performance of about the 35 th percentile of individual funds before fees. Thus, our approach provides a low-cost alternative to capturing the dynamic non-linear factor exposures of hedge funds. We show that most of the gains from hedge funds before and during the credit-crunch crisis could have been matched or even beaten by simply exploiting the information contained in publicly observable economic indicators, calling into question the issue of hedge fund manager skills.
Journal of Financial and Quantitative Analysis, 2012
In this paper we study the economic value and statistical significance of asset return predictabi... more In this paper we study the economic value and statistical significance of asset return predictability, based on a wide range of commonly used predictive variables. We assess the performance of dynamic, unconditionally efficient strategies, first studied by Hansen and Richard (1987) and Ferson and Siegel (2001), using a test that has both an intuitive economic interpretation and known statistical properties. We find that using the lagged term spread, credit spread, and inflation significantly improves the risk-return trade-off. Our strategies consistently outperform efficient buy-and-hold strategies, both in and out of sample, and they also incur lower transactions costs than traditional conditionally efficient strategies.
Uploads
Papers