配對交易是利用兩支具有長期均衡關係的股票，藉由價差的失衡獲取利潤的統計套利策略。當價差失衡時，交易者在表現較差的股票買空部位，在表現較好的股票賣空部位。如何選擇配對的股票與進出場的時機，是影響配對交易利潤的兩個重要因子。
本文利用共整合模型挑選配對的股票，藉由共整合係數去得到標準化的價差。我們對價差設定門檻，來決定進場的時機，當價差回到平衡點時就出場，藉此獲利並探討不同的門檻對獲利的影響。此外我們使用對數報酬率的平方建立配對股票波動的均衡模型，利用波動失衡的訊息，建立提早出場的機制。在實務操作時，我們利用固定時間長度的資料以動態的方式建立共整合模型及波動模型，每天更新模型來得到新的價差，並由設定的門檻決定是否進出場。
在實證分析時，我們對於S&P 500 中的四大類股去挑選出一萬多組具有共整合關係的配對組合，比較不同投資策略所賺取的利潤，及探討數種因子對利潤的影響。

Pairs trading is a statistical arbitrage strategy which gains profits via short-term
deviations from a long-run equilibrium between two stocks. Traders take a long position on underperforming stock, and a short position on outperforming one when the spreads loss equilibrium. Selection of the stock pairs and timing of entering a position are two important factors affecting the profits of pairs trading.
In this work, we use the cointegrated model to select the pairs of stocks, and use the cointegrated coefficient to obtain standardized spreads. We set spread thresholds to determine the timing of entering positions, and close the position to make a profit when the spreads return to balance. We explore the effect of the thresholds on the profits. In addition, we use the squared log returns to construct equilibrium models for stock volatility. Volatility imbalance are used to provide information for closing the position early. For practical implementation, we establish dynamic cointegration and volatility models using fixed-length data. We update the models daily to obtain new spreads, and enter a position when the spreads cross the set thresholds.
In empirical study, we selected around ten thousand sets of stock pairs from the four major sectors of S&P 500 which have cointegration relationship. We compare the profits of different trading strategies and explore the effect of several factors on profits.

誌謝........................................................................................ii
摘要 ...................................................................................... iii
Abstract ................................................................................ iv
1. Background and Motivation................................................1
2. Pair Selection.................................................................... .2
2.1 Mean Reversion............................................................2
2.2 Correlation.................................................................... 3
2.3 Cointegration.................................................................6
2.4 Construction of Cointegration Pairs...............................7
2.5 Standardized spreads...................................................10
3. Data Description..................................................................12
4. Trading Strategy .................................................................14
4.1 Strategies I & II..............................................................14
4.2 Strategies III & IV...........................................................14
5. Empirical Study....................................................................19

1. Chan, N.H. (2010). Time Series: Applications to Finance with R and S-Plus (2nd ed.). John Wiley & Sons, New Jersey.
2. Chiou, M. C. (2015). Analysis of Variables Affecting Pairs Trading Profits. Master thesis, Department of Applied Mathematics, National Sun Yat-sen University,
Kaohsiung.
3. Engle, R. and Granger, C. (1987) Co-integration and error correction: Representation, estimation, and testing, Econometrica, 55, pp. 251-276.
4. Gatev, E., Goetzmann, W. N., Rouwenhorst, K. G., 2006. Pairs Trading: Performance of a Relative-Value Arbitrage Rule. The Review of Financial Studies, 19(3), 797-827.
5. Johansen, Søren (1991). Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregressive Model. Econometrica, 59 (6): 1551-1580.
6. Johansen, S. and Juselius, K. (1990), Maximum likelihood estimation and inference on cointegration - with applications to the demand for money, Oxford Bulletin of Economics and Statistics, 52, 169-210.
7. Whistler, M.,2004. Trading Pairs - Capturing Profits and Hedging Risk with Statistical Arbitrage Strategies. (Wiley: New York).