我們使用紐約證券交易所的高頻交易資料，探討股票市場中具有影響力的交易。第一部份，我們將高頻交易資料，依據每筆交易資料對於市場長短期交易(如: 交易方向、報酬率以及交易量等變數)的影響，定義市場中具有影響力的交易。我們先將每一筆交易的與交易量同步的有資訊交易機率做排序並挑出前1000筆機率最高的交易，接著根據這些交易的報酬率、交易量以及交易方向作為變數，使用異常點偵測的方法將交易資料分成兩類 : 具有影響力交易以及一般交易。第二部份，我們考慮市場反應變數對具有影響力交易的反應變數建立羅吉斯回歸、支持向量機模型、異常點偵測技術以及最近鄰居法，並探討這四種方法對高頻交易資料類別的分類能力。

We use high frequency transaction data in NYSE, and investigate influential trade detection in stock markets. In part 1,
We defined the influential trade in stock markets with high frequency transaction data according to the effect of each short and long term transaction in market(such as: trade direction, return and volume etc.). We sort the volume-synchronized probability of informed(VPIN) decreasingly for each transactions and choose the transactions with the top 1000 probability. Then we regard the return, volume and trade direction rate for these transactions and use anomaly detection techniques to divide transaction data into two type : influential trades and ordinary liquidity trades. In part 2, we consider the four classified method(logistic regression, support vector machine, anomaly detection technique and $k$-nearest neighbors) and investigate the ability of classifying influential trades in high frequency transaction data.

Contents
誌謝i
摘要ii
Abstract iii
1 Introduction 1
2 Data Description 2
2.1 Matching Quotes and Trades . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Trade Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3 In
uential Trades 4
3.1 Probabilities of Informed Trading (PIN) . . . . . . . . . . . . . . . . . . . 4
3.2 Variables for Detecting In
uential Trades . . . . . . . . . . . . . . . . . . . 6
3.3 In
uential Trades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 Market Reaction Factors 9
4.1 Quoted Spread . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.2 Trade Prices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.3 Trade volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5 Methodology 12
5.1 Logistic Regression Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.2 Anomaly Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5.3 Support Vector Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5.4 K-Nearest Neighbors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
6 Empirical Studies 16
6.1 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
6.1.1 Logistic Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
6.1.2 Anomaly Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1.3 Support Vector Machine . . . . . . . . . . . . . . . . . . . . . . . . 19
6.1.4 K-nearest Neighbors . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.2 Summary of Blue Chip Stocks . . . . . . . . . . . . . . . . . . . . . . . . . 20
7 Conclusion 20
8 References 22
9 Appendix 24

[1] Barclay, M.J., and Warner, J.B. (1993). Stealth and Volatility: Which Trades Move
Prices? Journal of Financial Economics, 34, 281-305.
[2] Ben-Hur, A andWeston J. (2010) A User's Guide to Support Vector Machines. Methods
in Molecular Biology, 609, 223-239.
[3] Easley, D. and O'Hara, M. (1987). Price, Trade Size and Information in Securities
Markets. Journal of Financial Economics, 19, 69-90.
[4] Easley, D., Kiefer, N. M., O'Hara, M. and Paperman, J. B. (1996). Liquidity, information,
and infrequently traded stocks. Journal of Finance, 51, 1405-1436.
[5] Easley, D., Engle, R. F., O'Hara, M. and Wu, L., (2008). Time-Varying Arrival Rates
of Informed and Uninformed Traders. Journal of Financial Econometrics, 6, 171-207.
[6] Easley, D., Lopez de Prado, M. and O'Hara, M. (2011). Flow Toxicity and Liquidity
in a High-Frequency World. Review of Financial Studies, 25, 1457-1493.
[7] French, K.R. and Roll, R. (1986). Stock return variances: The arrival of information
and the reaction of traders. Journal of Financial Econometrics, 17,5-26.
[8] Glosten, L. and Milgrom, P. (1985). Bid, Ask and Transaction Prices in Specialist
Market with Heterogeneously Informed Traders. Journal Financial Economics, 14,
71-100.
[9] James, G., Witten, D., Hastie, T. and Tibshirani, R. (2015). An Introduction to Sta-
tistical Learning, New York: Springer.
[10] Jiang, J. (2015). Volume-Synchronized Probability of Informed Trading (VPIN), Market
Volatility, and High-Frequency Liquidity. MSc dissertation. Brock University:
Goodman School of Business.
[11] Lee, C. M. C., Ready, M. J. (1991). Inferring Trade Direction from Intraday Data.
Journal of Finance, 46, 733-746.
[12] Zagaglia, P. (2013). PIN: Measuring Asymmetric Information in Financial Markets
with R. The R Journal, 5, 80-87.