Thomas H. McInish

Cointegration, Error Correction, and Price Discovery on Informationally Linked Security Markets

Using synchronous transactions data for IBM from the New York, Pacific, and Midwest Stock Exchanges, we estimate an error correction model to investigate whether each of the exchanges is contributing to price discovery. Johansens test yields two cointegrating vectors, which together verify the expected long-run equilibrium of equal prices across the three exchanges. Two error correction terms specified as the differences from IBM prices on the NYSE indicate that adjustments maintaining the long-run cointegration equilibrium take place on all three exchanges. That is, IBM prices on the NYSE adjust toward IBM prices on the Midwest and Pacific Exchanges, just as Midwest and Pacific prices adjust to the NYSE.

Security price adjustment across exchanges: an investigation of common factor components for Dow stocks

Abstract VECMs can detect trades that permanently move the markets in cross-listed stocks. We employ Gonzalo and Grangers (J. Business Econom. Stat. 13 (1995) 1) reduced-rank regressions and QGG test statistic to analyze the common factor weight attributable to three informationally-linked exchanges for DJIA stocks over 1988–1995. We distinguish this error correction approach to trading price adjustment from the information shares approach to quote price leadership. In 1988, a 72.2% mean common factor weight (fNYSE) approximated the NYSEs 86% share of the trades. However, by 1992 fNYSE had declined precipitously for 27 Dow stocks, averaging only 49.6%, despite an unchanged 86% share of the trades. By 1995, the NYSEs common factor weight had recovered, averaging 62.9% on 84% of the trades. We discuss three alternative microstructure-theoretic hypotheses that can confront this evidence.

Individual investors and risk-taking☆

Abstract During the last decade, a large amount of information has been collected concerning financial markets and financial institutions. Less is known about individual investors. This study relates a specific personality characteristics, locus of control, to portfolio risk as measured by beta; these two concepts are described below. Locus of control and beta are examined in combination with the sex, marital status, age, educational level, asset level and number and value of common stocks held by one group of investors. The relationship between locus of control and beta is also considered. This study is divided into four parts. The first part reviews the capital asset pricing model and the construct “locus of control” as well as previous research relating locus of control to risky decision-making. Then, the data used in this study are discussed. The third section presents the results of this study, and the final part provides a summary.

An analysis of transactions data for the Toronto Stock Exchange : Return patterns and end-of-the-day effect

Abstract Using transactions data for all stocks traded on the Toronto Stock Exchange, this study shows that returns and number of shares traded have a U-shaped pattern when plotted against time of the trading day. These results confirm that the findings of Wood, Mclnish and Ord (1985), Harris (1986), Mclnish and Wood (1988) and Jain and Joh (1989) for the New York Stock Exchange (NYSE) also hold for both another exchange and another country and are not due to peculiarities of United States securities markets. Further, evidence is provided to support the view of Harris (1989) and Terry (1986) that these relatively high end-of-day returns are due, at least in part, to an increase in the proportion of trades at the ask relative to trades at the bid.

A transactions data analysis of the variability of common stock returns during 1980-1984

Abstract Using transactions data for a large sample of NYSE stocks for six months in 1971–1972 and calendar year 1982, Wood, McInish and Ord (1985) (WMO) show that the graph of the variability of index returns across days against time of day has a crude U-shaped pattern. This study demonstrates that relatively high variability of returns at the beginning and end of the trading day also occurs during calendar years 1980, 1981, 1983 and 1984. In addition, the variability of intra-minute returns across stocks is shown to have a crude U-shaped pattern when plotted against time of day. A model is developed and tested that explains this pattern of variability of intra-minute returns in terms of variability of market returns over the trading day. The empirical results are consistent with this explanation.

Production of information, information asymmetry, and the bid-ask spread: Empirical evidence from analysts' forecasts

In this paper we suggest that market makers deduce the extent of the adverse selection problem associated with a stock (and set up the bid-ask spread accordingly) by observing how many financial analysts are following that stock. Market makers do this based on the belief that more financial analysts would follow a stock with a greater extent of information asymmetry since the value of private information increases with informational asymmetry. Similarly, financial analysts deduce the profit potential of a stock from the size of the spread set up by market makers (based on the expectation that market makers would set up a greater spread for the stock with a greater information asymmetry). This structural view of the process determining the bid-ask spread and analyst following is empirically tested using a simultaneous equations regression analysis. The empirical results are generally consistent with this view. Hence, our study supports the notion that the decisions of two major players in the financial markets (i.e.. market makers and financial analysts) are made interactively.

An investigation of price discovery in informationally-linked markets: equity trading in Malaysia and Singapore

Abstract Using transactions data for the Kuala Lumpur Stock Exchange and the Stock Exchange of Singapore (SES) for a major Malaysian conglomerate, Sime Darby Berhad, and intraday exchange rate data, we investigate whether and to what extent each exchange contributes to price discovery. Results indicate that the price series are cointegrated. The raw data appear to indicate the presence of arbitrage opportunities, but none exist after taking exchange rate changes into account. Using the common long-memory factors of Gonzalo and Granger (1995, Journal of Business and Economic Statistics 13, 1–9), we show that while the majority of the price discovery (approximately 70%) occurs in the home country (Malaysia), the 26–32% of the price discovery attributable to the SES is statistically significant and exceeds Singapore’s share of the trading volume. Further, we find evidence of strong error correction of Singapore prices to Malaysian prices, but only weak error correction of Malaysian prices to Singapore prices.

Common factor components versus information shares: a reply ☆

Common factor components and information shares provide competing approaches to estimating the parameters of price discovery in cointegrated security markets or trading channels. Using simulated data on Hasbrouck’s (J. Financial Markets 5(3) (2002)) parameterized model of the stylized facts in satellite and centralized markets, we show that Gonzalo and Granger’s (J. Business Econom. Statist. 13 (1995) 1) procedure for estimating and testing common factor components recovers the true information structure in a wide range of financial market microstructure models. In addition, we investigate the role of stochastic process assumptions in estimating price discovery parameters by generalizing the sources of volatility in the implicit efficient price and the level of the signal to noise ratio. r 2002 Published by Elsevier Science B.V. JEL classification: G12

Reducing tick size on the Stock Exchange of Singapore

Abstract Minimum price changes, or tick sizes, are set by exchanges. Using theoretical and empirical models, previous researchers have concluded that transactions costs are increased unnecessarily if tick sizes are set higher than economic fundamentals justify. On July 18, 1994, the Stock Exchange of Singapore reduced the minimum tick size for stocks trading at

The Liquidity Of Automated Exchanges: New Evidence From German Bund Futures

25 or more from