Στοιχεία μικροδομής της αγοράς και η επίδρασή τους στην αποτελεσματική διαμόρφωση των τιμών

Abstract

This thesis examines market microstructure issues of the Greek stock market as regard two main issues: market efficiency in conjunction with the seasonality of stock pricing and closing price determination. Although studies based on intraday stock market indices and individual stock data are well-documented for the stock market in Greece, there is not any exploration of the microstructure effect upon the seasonality and closing price determination of the stock prices in Greece. This thesis investigates statistically the intraday behavior of the General index of the Athens Stock Exchange in order to discover possible intraday day patterns. We investigate for possible intraday patterns during different market phases, a mild “bull” and a “bear” period. In addition, we examine the possibility of intraday patterns during different days of the week. We obtained statistical evidence for a U shape pattern for stock returns and a measure of return volatility. Stock returns tend to be positive with the opening and the closing of a trading session and this pattern is statistically stronger on specific days of the week. The return U pattern formation is somehow different for a “bull” market as compared to a “bear” market. When intraday patterns are examined based on the day of the week, we have very strong evidence that the end of the trading period positive return is very significant for the case of Fridays especially in the “bull” market period, while for the opening price pattern, we observed significant positive returns on Wednesdays. We attribute these patterns to microstructure characteristics of the Greek market and to investors’ sentiment (emotional behavior). The intraday seasonality pattern of the Greek Market is an abnormality that could be attributed to market microstructure characteristics like market makers involvement in price formation process, pre and post trade information asymmetry between investors, short selling restraints, investors’ sentiment which is different during different market price movements (upward or downward). Further, we examine whether the abnormality defined in the closing prices is due to efforts from the participants to determine closing prices. We investigate the effectiveness of different methods in determining the closing price used in the Athens Stock Exchange (ASE) upon efforts of market participants to adjust closing prices at the desired level. We assess the transition from a value-weighted average price (VWAP) method to a plain-vanilla closing call auction method (CCAM). Our analysis provides strong evidence that the adoption of the closing call auction method improved substantially price formation but did not eliminate efforts of closing price determination. In addition, investors that trade during the closing price auction, take advantage of key elements of the trading system like the “reference price”, which may drive the closing price at the desired level. This thesis offers four main contributions to the literature. • First, our results indicate that intraday returns, together with the volatility of returns, follow a U-shaped pattern but the return pattern formation is different for a “bull” market that follows an upward trend which follows a U-shape pattern as compared to a “bear” market characterized by a downward trend in which there was no pattern existence. The results provide evidence that volatility is persistent for at least 30 min after the opening auction. All these can be attributed to the specific characteristics of the market microstructure. • Second, we assess the impact of a closing call auction method, and how a more elaborate version of it has affected the existence and magnitude of closing price adjustment. • Third, we introduce for the first time the notion of “strategic closing price handlers” vis-a-vis a daily closing price handler us we focus and use transactions of the dominant net buyers or sellers. In this way, we are in a better position to test the propositions of the Market Abuse Regulation directive considering the dominant position of the investors and how they adjust their trading strategy in marking the closing price. • Fourth, we examine the possibility that in a call auction setting, a trader may shift his trading practice just before the closing auction in order to influence the “reference price” to the desired level, which may, in turn, affect the closing price. The results can be useful to national regulators, exchanges, and researchers when they structure a stock market trading interface and a compelling closing price mechanism.