Font Size: a A A

Statistical Arbitrage Research Based On Program Trading

Posted on:2015-06-07Degree:MasterType:Thesis
Country:ChinaCandidate:J M ZhangFull Text:PDF
GTID:2309330431454774Subject:Applied statistics
Abstract/Summary:PDF Full Text Request
In the changable financial markets, the rational investment objective of market participants is to maximize efficiency under the expected risk. Since the global financial crisis in2008, the international financial market has un-dergone dramatic changes, resulting in huge losses of investors. However, s-tatistical arbitrage-based investment institutions are constantly refresh their own performance, fully demonstrates the advantages of statistical arbitrage trading, which is market neutral and can avoid systemic risk effectively. Sta-tistical arbitrage is basd on the historical price data of underlying investment assets, using statistical methods to control investment risk and gain market center revenue.The gain or loss of statistical arbitrage trading is little associated with the market trend. It mainly depends on the price relationship between the underlying portfolio. It needs perfectly short market mechanism. The paper studies statistical arbitrage strategies on the domestic futures market. With the stock index futures, margin financing and options markets continuously established and improved, statistical arbitrage strategy will get more applica-tion opportunities. In the mature capital markets like Europe and America, many large international fund companies and investment banks have already achieved excess return,using a lot of statistical arbitrage trading strategies on the program trading platform.Program trading does not pursuit good operation performance or exag-geration profits return under high risk conditions, it is prudent to pursue long- term profitability and growth in the market and achieve wealth accumulation compounding effect. After this kind of sound trading operations in a long period, the profit can be kept in a reasonable target level. It is characterized by more open trading strategy design, dynamic market risk management tech-niques and quickly issued feedback signal for testing the accuracy of fast error correction. After the transaction under a single transaction, it is conducive to the implementation of statistical arbitrage strategies in a large number of asset transactions and management. In contrast, program trading in the stock, fu-tures market is just domestically emerging applications. Because of the lack of short mechanisms, There are very few research and use of statistical arbitrage strategies.At present, the domestic research on statistical arbitrage trading gener-ally refers to the study of historical financial time series data, using statisti-cal models and methods to look for opportunities for arbitrage transactions. Traditional statistical arbitrage uses the experience and intuition in spreads between current and historical spreads deviation to judge trade opportuni-ties. The basic idea for the current research focus on statistical arbitrage and make historical data cointegration test to determine the presence of a se-quence of long-term stable relationship. If the long-term stable co-integration relationship between price series existed, then arbitrage opportunities could be interrelated developed.For example, if the price deviates from the equilibrium value, Transactions should be made; if the price return to the equilibrium value, profit would be attained. Cointegration arbitrage is essentially the use of mean reversion of the spread sequence. Mean reverting process is a spe-cial kind of random process, which is characterized by the gradual recovery of long-term average. In this paper, cointegration test result of the experimental data shows stable cointegration relationship, which means spreads have mean reversion characteristics.This paper assumes that the history of the spread sequence is a class of stochastic processes mean reversion. Innovation of this article is the use of modeling idea of program trading to calculate trade signal in the spread se- quence of dynamic processing, with fully use of the latest data. This paper selects an active Futures species, Shanghai Exchange rebar futures contract for the study. Domestic futures market now has a perfect short mechanism. statistical arbitrage trading ideas can be achieved there. While Chinese s-tock index futures, margin trading and options business gradually established, statistical arbitrage related research has forward-looking significance.
Keywords/Search Tags:Statistical Arbitrage, Program Trading, Cointegration, Trading Signals
PDF Full Text Request
Related items