Sampling interval and estimated betas: Implications for the presence of transitory components in stock prices

We provide a theoretical framework to explain the empirical finding that the estimated betas are sensitive to the sampling interval even when using continuously compounded returns. We suppose that stock prices have both permanent and transitory components. The discrete time representation of the beta depends on the sampling interval and two components labeled “permanent and transitory betas”. We show that if no transitory component is present in stock prices then no sampling interval effect occurs. However, the presence of a transitory component implies that the beta is an increasing (decreasing) function of the sampling interval for more (less) risky assets. In our framework, assets are labeled risky if their “permanent beta” is greater than their “transitory beta” and vice versa for less risky assets. Simulations show that our theoretical results provide good approximations for the estimated betas in small samples. We provide empirical evidence about the presence of negative serial correlation and mean reversion in the returns of the portfolios considered. We discuss why our model is better able to provide an explanation for this sampling interval effect than other models in the literature.     

A transactions data analysis of intraday betas

Using a sample of 27 stocks from the Dow Jones Industrial Average for the years 1986–1992, we examine the equality of beta for individual firms during the trading day. Both alphas and betas are found to differ through the trading day. Evidence suggests these changes are systematic for individual stocks. Using the midday beta as the base, the number of rejections of beta equality follow a U-shaped pattern through the trading day, indicating the differing distributions (U-shaped patterns) for intraday returns are reflected in similar changes in beta. These results have implications for further developing and testing market microstructure models.     

The 2000 presidential election and the information cost of sensitive versus non-sensitive S&P 500 stocks

We investigate the information cost of stock trading during the 2000 presidential election. We find that the uncertainty of the election induces information asymmetry of politically sensitive firms under the Bush/Gore platforms. The unusual delay in election results creates a significant increase in the adverse selection component of the trading cost of politically sensitive stocks. Cross-sectional variations in bid-ask spreads are significantly and positively related to changes in information cost, controlling for the effects of liquidity cost and stock characteristics. This empirical evidence is robust to different estimation methods.     

Semiparametric estimation of a characteristic-based factor model of common stock returns

We introduce an alternative version of the Fama–French three-factor model of stock returns together with a new estimation methodology. We assume that the factor betas in the model are smooth nonlinear functions of observed security characteristics. We develop an estimation procedure that combines nonparametric kernel methods for constructing mimicking portfolios with parametric nonlinear regression to estimate factor returns and factor betas simultaneously. The methodology is applied to US common stocks and the empirical findings compared to those of Fama and French.     

An encyclopedia for stock markets? Wikipedia searches and stock returns

We present empirical evidence that collective investor behavior can be inferred from large-scale Wikipedia search data for individual-level stocks. Drawing upon Shannon transfer entropy, a model-free measure that considers any kind of statistical dependence between two time series, we quantify the statistical information flow between daily company-specific Wikipedia searches and stock returns for a sample of 447 stocks from 2008 to 2017. The resulting stock-wise measures on information transmission are then used as a signal within a hypothetical trading strategy. The results evidence the predictive power of Wikipedia searches and are in line with the previously documented notion of buying pressure revealed by online investor attention and the trading patterns of retail investors.     

An Adjustment Procedure for Predicting Betas When Thin Trading is Present: Canadian Evidence

We have incorporated effects of the process that generates true betas for TSE stocks, as well as thin trading effects, into the beta adjustment model. We note the Blume and Dimson and Marsh beta adjustment techniques aim at eliminating beta forecast error through regression tendency bias. Effects of other sources of forecast error have been ignored. We show the process generating security betas affects both cross-sectional correlation coefficient and order bias, while thin trading affects only cross-sectional correlation coefficient. We demonstrate that when OLS beta estimates are used to forecast their future risk levels, order bias accounts for 86% of forecast error, while thin trading effects account for 14% of forecast error. A beta regression tendency model which properly accounts for effects of cross-sectional correlation (which is a function of thin trading) and order bias completely abates forecast error. Our results have implications for the use of correlation coefficient to measure stability of betas across time, for beta adjustment models proposed in the literature, and for event study methodologies that rely on prediction errors.     

Risk Interpretation of the CAPM's Beta: Evidence from a New Research Method

This study tests the validity of using the CAPM beta as a risk control in cross‐sectional accounting and finance research. We recognize that high‐risk stocks should experience either very good or very bad returns more frequently compared to low‐risk stocks, that is, high‐risk stocks should cluster in the tails of the cross‐sectional return distribution. Building on this intuition, we test the risk interpretation of the CAPM's beta by examining if high‐beta stocks are more likely than low‐beta stocks to experience either very high or very low returns. Our empirical results indicate that beta is a strong predictor of large positive and large negative returns, which confirms that beta is a valid empirical risk measure and that researchers should use beta as a risk control in empirical tests. Further, we show that because the relation between beta and returns is U‐shaped, that is, high betas predict both very high and very low returns, linear cross‐sectional regression models, for example, Fama–MacBeth regressions, will fail on average to reject the null hypothesis that beta does not capture risk. This result explains why previous studies find no significant cross‐sectional relation between beta and returns.     

THE EFFECT OF TRADING IN EIGHTHS ON SECURITY RETURN BEHAVIOR

Currently the equity securities of most British, Canadian and US firms trade in eighths. However, this pricing system may soon be abandoned in the US. Specifically, the US Securities and Exchange Commission (SEC) is currently studying the feasibility of changing the pricing of US securities to dollars and cents from dollars and eighths. 'SEC officials contend that moving to a system that quotes stock prices in dollars and cents would create efficiency in the stock market that eighths and sometimes sixteenths can't permit' (Torres and Salwen, 1991). This paper demonstrates the inefficiencies that result from constraining stocks to trade in eighths of a dollar. It describes the effects on returns and betas; then, it presents empirical evidence consistent with the effects. Systematic differences in the distributions of returns of low and high-priced stocks are documented. The covariance of returns with a market index is shown to vary systematically across stocks of different prices and to depend on the return interval used to estimate market model parameters. The variations are explainable by an observed lag between the returns of low-priced stocks relative to those of high-priced stocks. The lag is partially attributable to trading in eighths. A systematic relationship that varies with share price is observed between market model residual returns and unadjusted returns. This relationship is not eliminated by using longer return intervals alone. The extent of the relationship is reduced when longer return intervals are combined with the use of a market index composed of stocks that are priced similarly to those of the securities being tested. The implications of these results for capital market studies are discussed.     

S&P 500 trading strategies and stock betas

This paper shows that S&P 500 stock betas are overstatedand the non-S&P 500 stock betas are understated becauseof liquidity price effects caused by the S&P 500 tradingstrategies. The daily and weekly betas of stocks added to theS&P 500 index during 1985-1989 increase, on average, by0.211 and 0.130. The difference between monthly betas of otherwisesimilar S&P 500 and non-S&P 500 stocks also equals 0.125during this period. Some of these increases can be explainedby the reduced nonsynchroneity of S&P 500 stock prices,but the remaining increases are explained by the price pressureor excess volatility caused by the S&P 500 trading strategies.I estimate that the price pressures account for 8.5 percentof the total variance of daily returns of a value-weighted portfolioof NYSE/AMEX stocks. The negative own autocorrelations in S&P500 index returns and the negative cross autocorrelations betweenS&P 500 stock returns provide further evidence consistentwith the price pressure hypothesis.     

Information-based trading,price impact of trades,and trade autocorrelation

In this study we show that both the price impact of trades and serial correlation in trade direction are positively and significantly related to the probability of information-based trading (PIN). The positive relation remains significant even after controlling for the effects of stock attributes. Higher trading activity (i.e., shorter intervals between trades) induces both larger price impact and stronger positive serial correlation in trade direction. The effect of time interval between trades on quote revision is stronger for stocks with higher PIN values. These results provide direct empirical support for the information models of trade and quote revision.     

CAPM for estimating the cost of equity capital: Interpreting the empirical evidence

We argue that the empirical evidence against the capital asset pricing model (CAPM) based on stock returns does not invalidate its use for estimating the cost of capital for projects in making capital budgeting decisions. Because stocks are backed not only by projects in place, but also by the options to modify current projects and undertake new ones, the expected returns on stocks need not satisfy the CAPM even when expected returns of projects do. We provide empirical support for our arguments by developing a method for estimating firms' project CAPM betas and project returns. Our findings justify the continued use of the CAPM by firms in spite of the mounting evidence against it based on the cross section of stock returns.     

Delayed Reaction to Good News and the Cross-Autocorrelation of Portfolio Returns

We document a directional asymmetry in the small stock concurrent and lagged response to large stock movements. When returns on large stocks are negative, the concurrent beta for small stocks is high, but the lagged beta is insignificant. When returns on large stocks are positive, small stocks have small concurrent betas and very significant lagged betas. That is, the cross-autocorrelation puzzle documented by Lo and MacKinlay (1990a) is associated with a slow response by some small stocks to good, but not to bad, common news. Time series portfolio tests and cross-sectional tests of the delay for individual securities suggest that existing explanations of the cross-autocorrelation puzzle based on data mismeasurement, minor market imperfections, or time-varying risk premiums fail to capture the directional asymmetry in the data.     

The Price Is (Almost) Right

Most previous research tests market efficiency using average abnormal trading profits on dynamic trading strategies, and typically rejects the joint hypothesis of market efficiency and an asset pricing model. In contrast, we adopt the perspective of a buy‐and‐hold investor and examine stock price levels. For such an investor, the price level is more relevant than the short‐horizon expected return, and betas of cash flow fundamentals are more important than high‐frequency stock return betas. Our cross‐sectional tests suggest that there exist specifications in which differences in relative price levels of individual stocks can be largely explained by their fundamental betas.     

Direct Evidence of Non-trading on the London Stock Exchange

The extent of non-trading is shown to be much greater in the UK than in the more heavily researched US equity markets. Over the period 1975 to 1995 we find that almost 44% of all stocks in our sample failed to trade on the last day of a given month, a figure which is significantly higher than for stocks in the US (see Foerster and Keim, 1993). In this paper we investigate the relationship between the non-trading of UK stocks and the autoregressive and seasonal behaviour of UK stock returns. In addition, we find that stocks are much more likely to be recorded as not having traded on the last day of the month in the period prior to April 1981 than after this date. We trace this result to a reporting requirement change on the London Stock Exchange and investigate whether the change has any real implications for systematic risk estimates over this period. We also find that alternative methods for calculating betas, in the presence of thin trading, are very sensitive to stock size and to non-trading.     

RETURN INTERVAL,FIRM SIZE AND SYSTEMATIC RISK ON THE DUTCH STOCK MARKET

In this paper we investigate the behavior of betas of 50 Dutch firms as a function of the return measurement interval. We find beta estimates measured from different intervals differ significantly from each other. As the sample mainly contains stocks that are relatively thin compared to the index, beta estimates from short intervals are on average lower than those obtained from longer intervals. The results further indicate that there exists some variability in the beta coefficients for each interval length. Betas depend on the manner daily prices are juxtaposed to calculate the returns. A way to account for this variability is to average the different betas for each interval length. Asymptotic betas are also computed to show the appropriateness of this method. Finally we show that the size effect is reduced when the interval length is increased, although it remains statistically significant.     

中国A股市场大交易量与收益率关系的研究

利用技术分析制定股票投资策略是投资者主要采用的方法之一,而对交易量与收益率两者之间关系的研究又是技术分析的基础。我们认为,大交易量能更好地预测未来股票的收益。本文通过对中国A股市场代表不同规模股票的指数实证研究发现,不同指数在大交易量形成后的检验期里反应是不同的。代表大盘股的指数存在明显的"大交易量溢价效应";而代表小盘股的指数几乎不存在这种效应。我们还进一步的发现,这种"大交易量溢价效应"只发生在指数上涨了10%-20%的情况下。最后,我们给出了相关的投资策略。     

Factor beta,overnight and intraday expected returns in China

We study the relationship between common factor betas and the expected overnight versus intraday stock returns. Using data from the Chinese A-share markets, we find that the Fama-French five-factor betas and expected returns exhibit contrasting relationships overnight versus intraday. The market, value, and profitability factors earn positive beta premiums overnight and negative premiums intraday, while the size and investment factors' beta premiums behave oppositely. The night and day factor beta premium differentials are more muted among stocks with higher investor sophistication and vary across macroeconomic conditions. The contrasting day and night beta premiums extend to some other common factors and Chinese B shares, and vary their signs for some factors in the U.S. market.     

Option trading: Information or differences of opinion?

This paper investigates the motive of option trading. We show that option trading is mostly driven by differences of opinion, a finding different from the current literature that attempts to attribute option trading to information asymmetry. Our conclusion is based on three pieces of empirical evidence. First, option trading around earnings announcements is speculative in nature and mostly dominated by small, retail investors. Second, around earnings announcements, the pre-announcement abnormal turnovers of options seem to predict the post-announcement abnormal stock returns. However, once we control for the pre-announcement stock returns, the predictability completely disappears, implying that option traders simply take cues from the stock market and turn around to speculate in the options market. Third, cross-section and time-series regressions reveal that option trading is also significantly explained by differences of opinion. While informed trading is present in stocks, it is not detected in options.     

Volume autocorrelation,information, and investor trading

This study investigates whether the widely documented daily correlated trading volume of stocks is driven by individual investor trading, institutional trading, or both. We find that at least 95% of NYSE and AMEX stocks exhibit statistically significant, positive serial correlation. Volume autocorrelation decreases with the level of institutional ownership of a stock. We also show that the rate of arrivals of new information to the market contributes to the clustering of trades. When there is high information flow to the market, institutional trading generates a more pronounced effect on volume autocorrelation than individual investor trading. Our results are broadly consistent with the predictions of trading volume patterns suggested by most theoretical models of stock trading and by empirical research on investor trading.     

Adverse selection and the presence of informed trading

We combine two concepts of informed trading – contrarian trades and stealth trading – to develop proxies for the probability of informed trading. These proxies are used to test the link between informed trading and adverse selection as measured by bid–ask spreads and stock illiquidity. The estimation results show that these proxies, which are based on the probability of contrarian trading (PC) and progressively refined thereon, are all highly significantly positive in various empirical specifications of the cross-sectional determinants of spreads and illiquidity across stocks, and after controlling for important firm characteristics and trading factors. The robustness of our PC-based proxies for informed trading in these analyses, especially for the further refined measures, suggests that they successfully capture the adverse selection component of bid–ask spreads and illiquidity due to information asymmetry.