Optimal Pairs Trading Using Stochastic Control Approach---A Critical Evaluation
(2010) In MSc ThesesDepartment of Automatic Control
- Abstract
- As of date, quantitative trading methods are a strong growing niche of trading. More and more sophisticated models are employed to chase investment opportunities. The question is whether this is over-engineering and if it adds value to practitioners. We have found an interesting quantitative pairs trading strategy that belong to the family of relative value strategies. The studied stochastic control approach has many shortcomings. Among those is pair selection method which is not defined at all. Furthermore it has not yet been applied to real market data which makes it interesting to see how well it performs against a more basic pairs trading strategy. The purpose of this thesis is to define a suitable pair selection method that supports... (More)
- As of date, quantitative trading methods are a strong growing niche of trading. More and more sophisticated models are employed to chase investment opportunities. The question is whether this is over-engineering and if it adds value to practitioners. We have found an interesting quantitative pairs trading strategy that belong to the family of relative value strategies. The studied stochastic control approach has many shortcomings. Among those is pair selection method which is not defined at all. Furthermore it has not yet been applied to real market data which makes it interesting to see how well it performs against a more basic pairs trading strategy. The purpose of this thesis is to define a suitable pair selection method that supports the theoretical framework of stochastic control and to test this selection strategy against a more basic pairs trading strategy on historical market data. Through communication with hedge fund officials insight was gained on how investment strategies are applied in the market. Secondary data was gathered through the Datastream database, and the different simulations were run in MATLAB. Mathematical and economical theories were gathered through various textbooks and articles, and the direction of the study was discussed and decided with the advice of the supervisors. The method of stationarity through ADF test was found to be the best of the selection methods tested with the stochastic control approach. After finding proper time frames the stochastic control approach was benchmarked against a basic control strategy. The outcome shows that the specific quantitative strategy chosen for this study is flawed and therefore might not perform as well as it should with less assumptions made in the modeling. This is a sign of a possible over-engineering phenomenon that exists in the market in competition for investment opportunities. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8847462
- author
- Galeev, Roman and Ho, Wang Ching
- supervisor
- organization
- year
- 2010
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- Pairs Trading, Market Neutral Strategy, Stochastic Control, Cointegration, Hedge Fund Investment Strategies, Stationarity
- publication/series
- MSc Theses
- report number
- TFRT-5857
- ISSN
- 0280-5316
- language
- English
- id
- 8847462
- date added to LUP
- 2016-03-16 12:33:51
- date last changed
- 2016-03-16 12:33:51
@misc{8847462, abstract = {{As of date, quantitative trading methods are a strong growing niche of trading. More and more sophisticated models are employed to chase investment opportunities. The question is whether this is over-engineering and if it adds value to practitioners. We have found an interesting quantitative pairs trading strategy that belong to the family of relative value strategies. The studied stochastic control approach has many shortcomings. Among those is pair selection method which is not defined at all. Furthermore it has not yet been applied to real market data which makes it interesting to see how well it performs against a more basic pairs trading strategy. The purpose of this thesis is to define a suitable pair selection method that supports the theoretical framework of stochastic control and to test this selection strategy against a more basic pairs trading strategy on historical market data. Through communication with hedge fund officials insight was gained on how investment strategies are applied in the market. Secondary data was gathered through the Datastream database, and the different simulations were run in MATLAB. Mathematical and economical theories were gathered through various textbooks and articles, and the direction of the study was discussed and decided with the advice of the supervisors. The method of stationarity through ADF test was found to be the best of the selection methods tested with the stochastic control approach. After finding proper time frames the stochastic control approach was benchmarked against a basic control strategy. The outcome shows that the specific quantitative strategy chosen for this study is flawed and therefore might not perform as well as it should with less assumptions made in the modeling. This is a sign of a possible over-engineering phenomenon that exists in the market in competition for investment opportunities.}}, author = {{Galeev, Roman and Ho, Wang Ching}}, issn = {{0280-5316}}, language = {{eng}}, note = {{Student Paper}}, series = {{MSc Theses}}, title = {{Optimal Pairs Trading Using Stochastic Control Approach---A Critical Evaluation}}, year = {{2010}}, }