2019, Vol.7, No.1

Asset Management Review 2023 KCI Impact Factor : 0.55

pISSN : 2288-6672 / eISSN : 2713-6248
https://journal.kci.go.kr/capm

Most of portfolio performance measures used in the market do not consider higher order moment risk like skewness and kurtosis. This study presented general Sharpe ratio (GSR) considering higher order moment risk and estimated ex-post performance of zero cost investment strategies based on performance measures including GSR. It is summarized as follows. In the first place, cross-sectional correlations among Sharpe ratio, Treynor ratio, Jensen’s alpha, information ratio were found between 0.54 and 0.98. But correlation between GSR and these performance measures was located between 0.19 and 0.67. This difference is due to the fact that GSR considers higher order moments as additional risk factors. In the second place, investment strategy based on individual performance measure found statistically insignificant or negative performance. This supports the efficiency of KOSPI market and implies that it is hard to make a risk-adjusted excess return using performance measures.

The purpose of this study is to answer the question empirically whether we can expect model based strategy to deliver the additional risk-adjusted return and the reduction of risk using available information or not. Specifically we use Dynamic Nelson-Siegel model and its variants to derive expected returns and covariance for solving mean-variance optimization problem. Constructing bond portfolio using the optimal weight, we compare it with the traditional bond stagey and do model comparisons. We use Caldeira, Moura, and Santos (2016) approach to analyze KTB optimal portfolio construction. As the result of empirical analysis for the duration of January 2015 to December 2018, as risk appetite increases, it is efficient to move short-term investment to long-term as time goes by. Despite of the volatile out-of-sample period, model based strategy show a little lower but similar risk-adjusted performance.

A direct effect on the current performance of certain hedge fund investment styles by the one-month prior return of primitive trend following strategy on the short-term interest rate (PTFSIR) depends quadratically on the size of the returns of the one-month prior implied volatility index (“VIX”) as a primary moderator. To investigates empirically whether the past return of implied volatility information can explain the dynamic factor exposures of hedge funds over time, this manuscript advances the quantitative behavioral science literature for making inferences about the conditional process model with a quadratic moderator. Consistent with our view that hedge funds exhibit different levels of skills in exploiting the information contents of equity implied volatilities, we document substantial heterogeneity of processing one-month prior quadratic volatility returns by altering their risky asset exposures across hedge fund investment styles. We subsequently apply the same analytical framework to three different collections of individual liquid alternative hedge funds to test if the PTFSIR factor’s effect on hedge fund strategy returns is linearly moderated by one-month prior return volatilities of the VIX and the corresponding hedge funds’ dynamic risky asset and factor exposure management implications.