Kiseok Nam

ASYMMETRIC REVERTING BEHAVIOR OF SHORT HORIZON STOCK RETURNS: AN EVIDENCE OF STOCK MARKET OVERREACTION

Abstract This paper investigates the uneven mean reverting pattern of monthly return indexes of the NYSE, AMEX and NASDAQ, using asymmetric non-linear smooth-transition (ANST) GARCH models. It also evaluates the extent to which time-varying volatility in the index returns support the stock market overreaction hypothesis. The models illuminate patterns of asymmetric mean reversion and risk decimation. Between 1926:01 and l997:12, not only did negative returns reverse to positive returns quicker than positive returns reverted to negative ones, but negative returns, in fact, reduced risk premiums from predictable high volatility. The findings support the market overreaction hypotheses. The asymmetry is due to the mispricing behavior on the part of investors who overreact to certain market news. The findings also corroborate arguments for the “contrarian” portfolio strategy.

Asymmetric Nonlinear Smooth Transition Garch Models

Many studies of time varying volatility in stock markets have focused on an asymmetry in the volatility process known as “the leverage effect.” This asymmetry, first noted by Black in 1976, is characterized by the market’s asymmetric reaction to news, in the sense that stock market returns become more volatile after a negative price shock, than they do after a positive shock of the same absolute size. The first empirical models which explicitly attempted to capture the leverage effect include Engle’s (1990) A-GARCH model, Nelson’s (1991) E-GARCH model, and then the switching GARCH model proposed by Glosten, et al. (1993). Nowadays, there are many models which attempt to incorporate this asymmetry,1 and they are often conveniently summarized and compared using Engle and Ng’s (1993) “news impact curves,” which plot the predicted volatility response implied by a model, against past price shocks of given sign and size.

Asymmetric mean-reversion and contrarian profits: ANST-GARCH approach

Abstract This paper investigates the time-series evidence of asymmetric reverting patterns in stock returns that is attributable to “contrarian profitability.” Using asymmetric nonlinear smooth-transition (ANST) GARCH(M) models, we find that, for monthly excess returns of US market indexes over the period of 1926:01–1997:12, negative returns on average reverted more quickly, with a greater reverting magnitude, to positive returns than positive returns revert to negative returns. The results are quite consistent when the models are implemented not only for the different sample periods, such as 1926:01–1987:09 and 1947:01–1997:12, but also for portfolios with different characteristics, such as different firm-size portfolios and Fama–French risk-adjusted factor portfolios. We interpret the asymmetrical reversion as evidence of stock market overreaction.

Is asymmetric mean-reverting pattern in stock returns systematic? Evidence from Pacific-basin markets in the short-horizon

Abstract This paper applies asymmetric nonlinear smooth transition generalized autoregressive conditional heteroskedasticity (ANST-GARCH) models to the analysis of mean-reversion and time-varying volatility in weekly index returns of the stock markets of nine countries in the Pacific-basin. It finds that the returns exhibit an asymmetric pattern of return reversals, viz., on average, a negative return reverts more quickly, with a greater magnitude, to a positive return than a positive return reverting to a negative one. The asymmetric pattern of return reversals is directly associated with the unequal pricing behavior on the part of investors. Following a negative return shock, investors do not appear to require any additional premium to the leverage effect; instead they actually neutralize the risk in the form of a reduced premium! The reduction in risk premium causes not only the current stock price to rise but also the realized negative return to revert faster with a greater magnitude.

Profitability of Nonlinear Dynamics Under Technical Trading Rules: Evidence from Pacific Basin Stock Markets

This paper explores a possible link between an asymmetric dynamic process of stock returns and profitable technical trading rules. Using Pacific Basin stock market indexes, we show that the dynamic process of daily index returns is better characterized by nonlinearity arising from an asymmetric reverting property, and that the asymmetric reverting property of stock returns is exploitable in generating profitable buy and sell signals for technical trading rules. We show that the positive (negative) returns from buy (sell) signals are a consequence of trading rules that exploit the asymmetric dynamics of stock returns that revolve around positive (negative) unconditional mean returns under prior positive (negative) return patterns. Our results corroborate the arguments for the usefulness of technical analysis.

Revisiting the intertemporal risk-return relation: asymmetrical effect of unexpected volatility shocks

We suggest that an unexpected volatility shock is an important risk factor to induce the intertemporal relation, and the conflicting findings on the relation could be attributable to an omitting variable bias resulting from ignoring the effect of an unexpected volatility shock on the relation. With the effect of an unexpected volatility shock incorporated in estimation, we find a strong positive intertemporal relation for the US monthly excess returns for 1926:12–2008:12. We also find a significant link between the asymmetric mean-reversion and the intertemporal relation in that the quicker reversion of negative returns is attributed to the negative intertemporal relation under a prior negative return shock.

Flexible Multivariate GARCH Modeling for Heteroskedastic Asset Returns with an Asymmetric and Leptokurtic Distribution

We propose the S(subscript U)-normal distribution for the estimation of multivariate GARCH models to describe the nonnormality features, such as asymmetry and fat tails, embedded in heteroskedastic asset returns. We show that the S(subscript U)-normal distribution consistently outperforms the normal, Student-t and skewed-t distributions for describing the conditional distribution and the extreme lower and upper tail shapes of daily returns of individual stocks, industry portfolios, and national equity indexes over the sample period of 1989:01 - 2009:12. The exceeding ratio (ER) test for VaR forecasts suggested by Kupiec (1995) shows that the S(subscript U)-normal consistently outperforms the normal, Student-t, skewed-t distributions in multivariate CCC- and DCC-GARCH models. The results indicate that (a) compared to the S(subscript U)-normal, both the normal and Student-t distributions tend to underestimate the tail-thickness around the lower and upper extreme tails, and (b) even with an improvement relative to the normal and Student-t, the skewed-t distribution is still problematic because it tends to overestimate the extreme tails.