University of Fribourg (Switzerland) Faculty of Economics and Social Science Intraday Trading Activity on Financial Markets: The Swiss Evidence Thesis Submitted to the Faculty of Economics and Social Science of the University of Fribourg (Switzerland) in fulfillment of the requirements for the degree of Doctor of Economics and Social Science by Angelo Ranaldo from Sementina (TI) Accepted by the Faculty of Economics and Social Science on 17" February 2000 on the recommendation of Professor Jacques Pasquier-Dorthe (First Reporter) and Professor Nabil Khoury (Second Reporter) Fribourg, (Switzerland) 2000 Angelo Ranaldo was born in Sementina (Switzerland) on August 26, 1970. After attending scientific college in Bellinzona, he graduated in Business and Administration at the University L. Bocconi in Milan as «Dottore in «La Faculté des sciences économiques et Economia e Commercio». During his Ph., he sociales de JVuniversité de Fribourg worked at the University of Fribourg and (Suisse) n’entend ni approuver, nỉ attended the Gerzensee Ph.
program run by the désapprouver les opinions émises dans une Swiss National Bank. Currently, he is a Visiting thése: elles doivent étre considérées Scholar at the New York University, Stern comme propres a l’auteur (Décision du School of Business as a post-doctorate researcher Conseil de Faculté du 23 janvier 1990». in the Finance Department. ACKNOWLEDGMENTS I carried out my Ph.
thesis while working at the University of Fribourg (CH). During this period the chair of Finance was directed by Professor J.Pasquier-Dorthe and the working team consisted of Dr S. Gay Robin, Dr R. Haberle and later M.
I was very fortunate to be part of this team for two main reasons: first, because of the stimulating collaboration from which I learnt enormously and, secondly, because the team created an environment in which human aspects always took first place. In particular, I owe a debt to Prof. Pasquier-Dorthe who always showed immense sensitivity and understanding. He gave me the opportunity to attend the full Ph.
program in Gerzensee (1997-8), he provided helpful feed-back on my work and, more important, he always gave importance to our human relationship. I am also very grateful to Sophie Gay Robin for her friendship and her useful advice. I also thank all the people who made helpful suggestions related to my Ph. In particular, I would like to acknowledge Prof.
Khoury who undertook to supervise my dissertation. I also thank Prof. Robert Engle (UCSD), Prof. Joel Hasbrouck (NYU), Prof.
Bo Honoré (Princeton University) and Prof. Gouriéroux (CREST) and many colleagues at the University of Fribourg such as C. I am extremely grateful to the Swiss Stock Exchange, in particular to J. Wick, who graciously provided the dataset.
I want also to express my gratitude to my family and to all those who gave me moral support. First of all I want to thank my partner, Karin, who is the main source of my sentiments and commitments. I will never thank her enough for her love for me. I am also especially indebted to my mother who provided continuous and generous attention.
When I think of these two and of all my good friends I recognize my boundless fortune. As regards my friends I would not forget the “bohemian clan’, namely Chico, Diego, Fulvio, Gianlu, Gigi, Gio, Johnny, Marco, Max, Piffo, Rocco, and others such as Alberto, Andrea, Angelo, Michi, Mario and Omar. A final thought is reserved for the memory of my grandfather Rocco to whom this work is dedicated.1 Market Structures, oo.2 Microstructure MOd€ÏS,.3 High-Frequency IDAfA,. -«s«s=xxseexseexsesexseexsesrxsi 25 1: INTRADAY MARKET LIQUIDITY, Q22 xnxx xe, 31 1.
Description of the Market and Dataset. Determinants of Market Liquidity. CONCLUSION, LL esseeeeessseeeensseesenseeeeenneeeeenseeeeeeseeeeess 63 ¬. - - se set XE TH HH HT xe 79 2: THE INFORMATION CONTENT OF ORDER VOLUMES _ 87 PC on nh.
Background and Literature Review, 92 Intraday Trading Activity on Financial Markets 2. Description of the Market and Dataset. LIST OF ABREVIATIONS 2.4 The Tick-By-Tick Relationships AC: Auto Correlation ACD: Auto Conditional Duration Adj. R-2: Adjusted R-squared AGEFI: a Newspaper of the French Swiss AIC: Akaike Information Criterion APT: Asset Pricing Theory 3: LEAD-LAG RELATIONSHIPS BETWEEN STOCKS ARCH: Auto Regressive Conditional AND OPTIONS 131 Heteroskedasticity 3.
133 ARMA: Auto Regressive Moving Average 3. Introduction 134 CAPM: Capital Asset Pricing Model 3. Review of the Literature 136 CATS: Computer Aiding Trading System 3. Dataset, Market Structure and Methodology, 142 CBOE: Chicago Board of Exchange 3.: Durbin- Watson Statistic 3.
Conclusion 155 FR: Flow Ratio 3. Figures 157 GARCH: Generalized ARCH 3. Tables 159 Log likelL.: Logarithmic Likelihood LR: Liquidity Ratio 4. CONCLUSIONS 165 LSE: London Stock Exchange 4.
Intraday Market Liquidity, 167 NASDAQ: National Association of Securities Dealers Automated Quotations 4. The Information Content of Order Volumes 174 NYSE: New York Stock Exchange 4. Lead-Lag Relationships between Stocks and Options,,_. 177 NZZ: Neue Zuercher Zeitung 4.
Research Agenda 181 OR: Order Ratio 5. REFERENCES 183 OWAIT: the Waiting Time between the Time Arrival of Two Subsequent Orders PAC: Partial Auto Correlation Prob(F-s): Probability related to the F-Statistic PSE: Paris Stock Exchange Intraday Trading Activity on Financial Markets List of Abbreviations RBSVI: Ratio of Volume Imbalance between the VIMB: Order Volume Imbalance between the Buy and the Sell Part of the Market Buy and the Sell Part of the Market RGINI: Ratio of the Gini Index VIMBAV: VIMB in Absolute Value RLRC: Ratio of the First Level of the Return VPUT: Cumulated Trading Volumes of Options Autocorrelation RS: Ratio of Bid-Ask Spread VR: Variance Ratio RTAV: Ratio of Trading Volume Average VT: Trading Volume RTV: Ratio of Trading Volume WT: Waiting Time between Subsequent Trades RVR: Ratio of Returns Volatility S.: Standard Deviation of Dependant Variable SEAQ: Stock Exchange Automated Quotation System S.: Standard Error of the Regression SMI: Swiss Market Index SOFFEX: Swiss Options and Financial Futures Exchange SPI: Swiss Performance Index SRETURN: Stock Return SSR: Sum of Squared Residuals SVOL: Cumulated Trading Volume on Stock Market SWAIT: the Mean of the Waiting Time between Subsequent Trades SWX: Swiss Stock Exchange TARCH: Threshold ARCH URVT: Ratio of Unexpected Trading Volume VAR: Vector Auto Regression VCALL: Cumulated Trading Volumes of Call Options VCP: Cumulated Trading Volumes of Call and Put Options 10 II LIST OF TABLES Introduction Table 0.1: The Market Structures of the Main Stock Markets in the World by Agency and Dealer Markets, by Continuous and Call Markets. 20 1: Intraday Market Liquidity Table 1.1: The Pearson Correlation between Eight Liquidity Proxies 69 Table 1.2: Fifteen Swiss Stocks as Ranked by Different Liquidity Proxies 70 Table 1.3: An Estimation of Intraday Market Concentration,,__.4: Intraday Market Depth as Trading Volume,__.5: Intraday Market Depth Estimated by Order Volume Tmbalance ,.6: Time Dimension of Intraday Market Liquidity.7: Tightness of Intraday Market liquidity, 75 Table 1.8: Intraday Relationships between Spread and Trading `".9: Intraday Return VOIAaHÏIV,.-- «ssxs=exsesexsersee 77 Appendix 1.1: Proxies of Intraday Market Liquidity,,.2: The Gini Ïh€XL.«- cs xxx xeesessze 82 Appendix 1.3: Intraday Market VariabÏ©9$,.4: The Distribution of the Four Cases 85 2: The Information Content of Order Volumes Table 2.1: Tick-by-Tick Relations between Volume Imbalances and Returns the Fifteen Swiss Stocks 113 12 13 Intraday Trading Activity on Financial Markets Table 2.2: Tick-by-Tick Relationships between Order Volume Imbalances and the Waiting Time between Orders for the Fifteen Swiss Stocks 115 Table 2.3: Tick-by-Tick Relationships between Order Volume Imbalances and Returns over the Trading Day for the Novartis and the Nestle Stocks 117 Table 2.4: Tick-by-Tick Relationships between Order Volume Imbalances and Returns over the Trading Day for the UBS N and the Clariant Stocks 119 Table 2.5: Tick-by-tick Ordered Probit Model applied to Fifteen Swiss Stocks 121 Table 2.6: The Ordered Probit Model over the Trading Day: the Novartis Stock 123 Appendix 2.1: The Distribution of the Ten Intraday Events_, 125 Appendix 2.2: The Ordered Probit Model over the Trading Day: the CS Stock 126 Introduction Appendix 2.3: The Ordered Probit Model over the Trading Day: the Clariant Stock 128 3: Lead-Lag Relationships between Stocks and Options Table 3.1: Intraday Relationships between Option and Stock Volumes 159 Table 3.3: Waiting Time to Trade on the Stock Market.4: Intraday Relationships between Call and Put Option Volumes, and Waiting Time to Trade on the Stock MMATKC Q00 xi Hy HH xen sen 162 Table 3.5: Intraday Relationships between Option Volumes and Stock Returns 222222 xnxx BeXn vn xen ng xen 163 Table 3.6: Granger Causality Test Resuls_. 164 14 15 Introduction Abstract This study is a theoretical and empirical research on financial markets.
In particular, we focus on microstructure theory and intraday empirical investigations, which are two of the most recent developments in Finance. The empirical analysis is based on a high-frequency dataset of Swiss stock and option markets. The importance of these research areas has several roots. Historically, since the beginning of the ‘80s a large number of financial markets around the world have been changing their structures and have become informatized.
Practically, markets are more and more inter-linked and traders take intraday positions. The organization of the introduction is as follows.1 we try to describe the historical evolution and the main features of market structures.2 is a survey of the microstructure literature while Section 0.3 emphasizes the most important outlines of the research areas based on high frequency data. In order to help the reader, along this introduction we will write in italic the original contributions presented in the other parts of this study. 16 17 Intraday Trading Activity on Financial Markets Introduction 0.
MARKET STRUCTURES and the sell order are directly matched while a price driven market is an exchange system where the traders must trade with a market- The structure of a securities market refers to the systems, maker who continuously provides a bid and an ask price (see, for procedures and rules that determine how orders are handled and example, the NASDAQ and the SEAQ). In some markets the market translated into trades and how transaction prices are set. From this maker is the monopolist for a given asset, as on the NYSE where he point of view, the micro-foundation of financial analysis is is called “the specialist’, while in many other cases market makers enormously important. While much of economics is concerned with are in competition.
the trading of assets, market microstructure research focuses on the The third criterion is based on the trading space, which can interaction between the mechanics of the trading process and its be centralized or fragmented. A trading system is_ spatially outcomes, with the specific goal of understanding how actual fragmented if orders can be routed through different markets. There markets and market intermediaries behave (Easely and O’Hara, are many types of market fragmentation: order flow may be 1995). This focus allows researchers to pose applied questions fragmented for exchange listed issues and issues may be cross-listed regarding the performance of specific market structures, as well as (listed on more than one exchange); some orders are handled more theoretical queries into the nature of price adjustment.
differently from other orders (for instance small orders are routed to The preliminary task of this introduction is to briefly define immediate execution or large block trades are negotiated off-board in the main features characterizing a financial market. Following the an upstairs market). framework of Biais, Foucaoult and Hillon (1997) we shall use three Much electronic equipment has been introduced in recent principal criteria to classify the different typology of market years. Since Toronto became the first stock exchange to computerize structures: (1) the trading time, (2) the market agents, and (3) the its execution system in 1977, electronic trading has been instituted in trading place.
Tokyo (1982), Paris (1986), Australia (1990), Germany (1991), As regards trading time we distinguish between continuous Israel (1991) Mexico (1993), Switzerland (1995), and elsewhere versus call markets. A continuous market allows trades to be made at around the globe. In computerized trading, orders electronically any time during a trading day that counterpart orders cross in price.