An Analysis of Herding in Korean Stock Market Using Network Theory

Soosung Hwang,Young-Il Kim,Jinho Shin 한국재무학회 2016 한국재무학회 학술대회 Vol.2016 No.05

We investigate whether herd behavior in equity market is led by ‘core’ stocks or by ‘peripheral’ stocks connected to core stocks, which we identify using the minimum spanning tree, a technique in network theory. Using non-securities stocks listed in the Korea Exchange from January 2005 to December 2015, we find that core stocks are not necessarily the stocks whose market values are large but are mid-sized stocks. As in previous studies, we find strong evidence of herding in the Korean stock market. However, herding arises only when the market is in stress: during bear states, core stocks herd toward to the market return and peripheral stocks herd to core stocks in their clusters. During bull markets, however, adverse herding arises mainly driven by securities stocks and thus cross-sectional dispersion in returns increases.

The Disappearance of Momentum

Soosung Hwang,Alexandre Rubesam 한국재무학회 2011 한국재무학회 학술대회 Vol.2011 No.05

We investigate the robustness of the momentum premium in the US over the period from 1927 to 2010 using a model that allows multiple structural breaks. We find that the risk-adjusted momentum premium is significantly positive only during certain periods, notably from the 1940s to the mid-1960s and from the mid-1970s to the late 1990s, and we find evidence that momentum has disappeared since the late 1990s. Our results further suggest that momentum profits have been slowly eroded away since the early 1990s, in a process which was delayed by the occurrence of the high-technology stock bubble of the 1990s. In particular, we estimate that the bubble accounted for at least 50% of momentum profits during the period from 1994 to 2000.

Overconfidence and Subsequent Return Reversals

Soosung Hwang 한국재무학회 2015 한국재무학회 학술대회 Vol.2015 No.05

I investigate investors’ response to market-wide and firm-specific signals during their Bayesian updating process by observing cross-sectional return patterns. My empirical results show that investors’ overconfidence, that is, an over-response to a signal, is more likely to occur for mature firms that are relatively easy to price: large firms, value firms, dividend-paying firms, firms with more tangible assets, firms with little external financing, and firms with low sales growth. The characteristics of these firms stand in sharp contrast with those of firms that are more likely to be affected by sentiment. I also find that the effects of investors’ overconfidence on returns are reversed immediately. Therefore, the over-precision in the market-wide and firm-specific signals that are investigated in this study is not responsible for various anomalies, such as the value premium that necessitates a delay in the response or reversals over longer horizons. Instead the unexpected response to a signal by overconfident investors explains a significant proportion of short-term return reversals.

An Analysis of Herding in the Korean Stock Market Using Network Theory

Soosung Hwang(황수성),Young-Il Kim(김영일),Jinho Shin(신진호) 한국증권학회 2018 한국증권학회지 Vol.47 No.3

본 연구에서는 주식시장의 스트레스 상황 하에서 투자자들이 기업의 본질가치 변화나 소속된 산업의 움직임 보다는 네트워크 상에서 밀접하게 ‘연결된 주식’의 가격변화를 추종하는지 여부를 네트워크 이론으로 분석한다. 실증분석 결과, 2005년 1월부터 2015년 12월까지의 기간 동안 한국주식시장에서 군집행동이 나타났으며, 이는 기존의 문헌과도 일치하는 결과이다. 흥미로운 점은 시장이 불황기일 때 군집행동은 주가가 상승하는 경우에 나타난다는 것이다. 이러한 결과는 과신하는 투자자들이 가격 상승이라는 긍정적인 신호에 대해 과민반응하는 것으로 해석할 수 있다. 왜냐하면, 시장이 불황일 때는 현재시점의 가격이 낮기 때문에 투자자의 기대수익률이 높은데, 이 때 긍정적인 신호가 나타난다면 이 신호는 높은 수익률을 기대했던 투자자들의 사전적인 믿음과 일치하는 것이므로 자기귀인편향(self-attribution bias)을 유발시키기 때문이다. 반면, 시장 불황기에서 주가가 급락하는 경우에는 군집행동이 항상 나타나지는 않았다. Using network theory, we investigate if investors follow movements of closely ‘connected stocks’ regardless of their fundamentals or industries when investment decision is driven by panic under stress. We find strong evidence of herding in the Korean market for the period from January 2005 to December 2015 as in previous studies in herding. However, herding arises at positive extreme market movements during bear states. We interpret the results as follows: during bear states when the expected market return is high, overconfident investors over-respond to good signals because the signals are consistent with their priors of the high expected market return (self-attribution bias). Herding does not necessarily arise by panic under stress.

Sentiment, Beta Herding, and Cross-sectional Asset Returns

Soosung Hwang,Mark Salmon 한국재무학회 2013 한국재무학회 학술대회 Vol.2013 No.05

We investigate the effects of a behavioral bias in betas on cross-sectional asset returns. This bias, which we call beta herding, reflects the interaction between sentiment and herding in linear factor models. We demonstrate that beta herding is likely to arise when investors are more confident about the future direction of the market, regardless of whether the market is rising or falling, or when sentiment is high. Contrary to common belief that herding increases during market crises, our empirical evidence indicates that crises appear to lead investors to seek out the fundamental risk-return relationship rather than to herd. In terms of cross-sectional asset returns, we show evidence that beta matters conditionally on beta herding though it does not unconditionally. The conditional explanation of beta herding is distinct from other firm characteristics.

Overconfidence and Excess Arbitrage

Min Hwang,Soosung Hwang,Sanha Noh 한국재무학회 2013 한국재무학회 학술대회 Vol.2013 No.05

We investigate the effects of arbitrageurs’ behavioral biases on cross-sectional equity returns under the assumption that arbitrageurs are Bayesian optimizers. We demonstrate that the profits of equity market neutral trading strategies are positively affected by overestimation, but do not find evidence of overprecision and self-attribution bias in these trading strategies. Of the last four decades since 1970, the effects of overconfidence on the profits of the trading strategies are the highest in the 2000s when arbitrage trading is active. Despite the active arbitrage trading, however, the potential profitability measured by alphas does not seem to be eroded away. As an explanation of why the performance of equity market neutral hedge portfolios appears to decrease significantly in the 2000s, we propose excess arbitrage trading driven by overconfidence. That is, the temporal profits of arbitrage strategies temporally increased by excessive trading of overconfident arbitrageurs are subsequently reversed. The poor performance that hedge fund managers suffer comes from the reversal of the selected hedge portfolios, the selection of which is affected by upward bias in temporal profits.

Ambiguity, Overconfidence, and Excess Arbitrage

Min Hwang,Soosung Hwang,Sanha Noh 한국재무학회 2012 한국재무학회 학술대회 Vol.2012 No.05

We investigate the effects of arbitrageurs’ behavioral biases on cross-sectional equity returns under the assumption that arbitrageurs are Bayesian optimizers. We find that the profits of various equity neutral hedge portfolios are affected by arbitrageurs’ attitude toward ambiguity to signals. Ambiguity-aversion (Epstein and Schneider, 2008) exists before 2000 for some hedge portfolios whose idiosyncratic volatilities increase at positive signals since the positive signals are interpreted as being ambiguous. However, during the 2000s arbitrageurs interpret positive signals as being clear rather than ambiguous, because positive signals confirm their prior beliefs that the trading strategies are profitable. The excess arbitrage from the misinterpretation of positive signals explains why the profitability of equity neutral hedge portfolios appears to decrease in the 2000s despite the fact that their alphas are still positive and significant.