Soosung Hwang

Modelling Emerging Market Risk Premia Using Higher Moments

The purpose of this paper is to assess the incremental value of higher moments in modelling capital asset pricing models (CAPMs) of emerging markets. Whilst it is recognized that emerging markets are unlikely to yield sensible results in a mean-variance world, the high skewness and kurtosis present in emerging markets returns make our assessment potentially interesting. Generalized method of moments (GMM) is used for the estimation. We also present new versions of higher-moment market models of the data-generating process of the individual emerging markets and use these to identify model parameters. We find some evidence that emerging markets are better explained with additional systematic risks, such as co-skewness and co-kurtosis, than the conventional mean-variance CAPM. Copyright @ 1999 by John Wiley & Sons, Ltd. All rights reserved.

Market risk and the concept of fundamental volatility : measuring volatility across asset and derivative markets and testing for the impact of derivatives markets on financial markets

This paper proposes the unobserved fundamental component of volatility as a measure of risk. This concept of fundamental volatility may be more meaningful than observed volatility for market regulators. Fundamental volatility may be obtained using a stochastic volatility model. The authors decompose four FTSE100 stock index related volatilities into transitory noise and unobserved fundamental volatility. The question as to whether derivative markets destabilise asset markets is addressed. The analysis shows that introducing European options reduces fun-damental volatility, while transitory noise in the underlying and futures markets does not show significant change. It is concluded that, for the FTSE100 index, introducing an options market has stabilised underlying and derivative markets.

Small sample properties of GARCH estimates and persistence

Abstract It is shown that the ML estimates of the popular GARCH(1,1) model are significantly negatively biased in small samples and that in many cases converged estimates are not possible with Bollerslev’s non-negativity conditions. Results also indicate that a high level of persistence in GARCH(1,1) models obtained using a large number of observations has autocorrelations lower than these ML estimates suggest in small samples. Considering the size of biases and convergence errors, it is proposed that at least 250 observations are needed for ARCH(1) models and 500 observations for GARCH(1,1) models. A simple measure of how much GARCH conditional volatility explains squared returns is proposed. The measure indicates that for a typical index return volatility whose ARCH parameter is very small, the conditional volatility hardly explains squared returns.

The Effects of Systematic Sampling and Temporal Aggregation on Discrete Time Long Memory Processes and their Finite Sample Properties

This study investigates the effects of varying sampling intervals on the long memory characteristics of certain stochastic processes. We find that although different sampling intervals do not affect the decay rate of discrete time long memory autocorrelation functions in large lags, the autocorrelation functions in short lags are affected significantly. The level of the autocorrelation functions moves upward for temporally aggregated processes and downward for systematically sampled processes, and these effects result in a bias in the long memory parameter. For the ARFIMA(0,d,0) process, the absolute magnitude of the long memory parameter, |d|, of the temporally aggregated process is greater than the |d| of the true process, which is greater than the |d| of the systematically sampled process. We also find that the true long memory parameter can be obtained if we use a decay rate that is not affected by different sampling intervals.

GARCH model with cross-sectional volatility: GARCHX models

This study introduces GARCH models with cross-sectional market volatility, which are called GARCHX models. The cross-sectional market volatility is a special case of common heteroscedasticity in asset specific returns, which is suggested by Connor and Linton (2001) as an important component in individual asset volatility. Using UK and US data, we find that daily return volatility can be better specified with GARCHX models, but GARCHX models do not necessarily perform better than conventional GARCH models in forecasting.

An Analysis of Performance Measures Using Copulae

Publisher Summary There are several issues of immediate importance to evaluate performance of a portfolio although the relationship between the risk carried by the portfolio and the return is clearly paramount. Different concepts of risk, either relative or absolute, are employed and the notion of risk itself may be measured more generally using quantiles than the traditional use of the second moment. Another issue is the measurement of any superior timing ability as distinct from the stock selection ability in the fund manager. The difference between these two turns on whether private information lies in market aggregates or is firm specific. Portfolio returns are invariably not normally distributed and higher moments such as skewness and kurtosis need to be considered to adjust for the non normality and to a degree account for the failure of variance to measure risk accurately. In these cases, a higher moment capital asses pricing model (CAPM) should prove more suitable than the traditional CAPM and so a performance measure based on higher moments may also be more accurate than other measures.

Smoothing, Nonsynchronous Appraisal and Cross-Sectional Aggregation in Real Estate Price Indices

In this article three econometric issues related to private-equity return indices, such as real estate indices, are explored (smoothing, nonsynchronous appraisal and cross-sectional aggregation). Under certain assumptions, it is found that index returns based on appraisals follow an ARFIMA(1, d, 1) (autoregressive fractionally integrated moving average) process, where the long memory parameter (d) explains the level of smoothing and the AR and MA parameters represent the level of persistence in marketwide fundamentals and the nonsynchronous appraisal, respectively. The empirical results show that: (1) the level of smoothing in appraisal-based real estate indices is far less than assumed in many academic studies (2) there is weak evidence of nonsynchronous appraisal in the UK, IPD (Investment Property Databank) index and (3) marketwide fundamentals are highly persistent for the IPD index returns. On the other hand, there is no evidence of nonsynchronous appraisal or a persistent common factor in the U.S. NCREIF (National Council of Real Estate Investment Fiduciaries) index.

11 – Implied volatility forecasting: a comparison of different procedures including fractionally integrated models with appications to UK equity options

The purpose of this paper is to consider how to forecast implied volatility for a selection of UK companies with traded options on their stocks. The authors consider a range of GARCH and log--ARFIMA based models as well as some simple forecasting models. Overall, it is found that a log-ARFIMA model forecasts best of short and long horizons.

Does downside beta matter in asset pricing

By carefully choosing a data-generating process and appropriate distributional assumptions, we formulate a nested econometric model to examine how many equities are explained well by the downside beta or a general asymmetric response model rather than the conventional capital asset pricing model (CAPM) beta. Using UK equity data, we show that the downside beta explains 15–25% of equities in addition to CAPM that explains 50–80% of equities. These results suggest that although the lower partial moment CAPM explains equity returns better than the conventional CAPM, the proportion of equities benefiting from using the downside beta is not large enough to improve asset pricing models significantly.

Calculating the misspecification in beta from using a proxy for the market portfolio

This study investigates the effects of the market portfolio being unknown on the estimation of beta in the CAPM. Providing an analysis of the impact of using a proxy for the market portfolio when the market portfolio is known. This allows one to ask and answer ‘if what’ questions, such as if portfolio A is the true market portfolio, what happens to beta if one uses portfolio B as a proxy for A. It is shown that for a given universe of investible assets, frequently used equally weighted and value weighted portfolios are far from the Markowitz market portfolio and thus the betas calculated with the equally weighted and value weighted portfolios are quite different from those obtained with the Markowitz portfolio. These calculations are based on sequential assumptions that one portfolio is a proxy whilst another is the actual market.