【摘要】：獲得超額回報(bào)——這一直以來都是投身金融交易的廣大投資者的終極追求,而有效而又科學(xué)的進(jìn)行交易,就必須依托于一套有效的交易策略,無論是基本面分析還是技術(shù)分析,從本質(zhì)來講,都有其內(nèi)在的策略。 程序化交易就是將投資的策略進(jìn)行量化處理的過程,設(shè)計(jì)的策略由計(jì)算機(jī)識(shí)別并執(zhí)行,在國外發(fā)達(dá)市場中,無論是對(duì)個(gè)人投資者還是對(duì)于大型機(jī)構(gòu)投資者,程序化交易都是最前沿、最科學(xué)的投資模式。程序化交易可以在復(fù)雜和無序的價(jià)格波動(dòng)中,尋求非隨機(jī)的獲利機(jī)會(huì)。這種非隨機(jī)的獲利機(jī)會(huì),表現(xiàn)為市場價(jià)格運(yùn)動(dòng)存在記憶性,即價(jià)格通常會(huì)持續(xù)沿著一個(gè)方向進(jìn)行運(yùn)動(dòng),直至本階段行情結(jié)束,一旦方向被改變,價(jià)格運(yùn)動(dòng)又會(huì)沿著另一個(gè)方向繼續(xù)持續(xù)。這種記憶性特征被人們定義為趨勢,同時(shí),趨勢的存在為程序化交易提供了設(shè)計(jì)的方向。 我國目前關(guān)于程序化交易模型設(shè)計(jì)的文章近乎空白。與交易策略有關(guān)系的文章所涉及的策略大都很簡單,比如普通的均線穿越系統(tǒng)、MACD、KDJ等經(jīng)典指標(biāo)衍生出來的交易系統(tǒng)。這些檢驗(yàn)大多是驗(yàn)證其正回報(bào)特點(diǎn),而對(duì)于系統(tǒng)資金曲線穩(wěn)定程度、資金回撤率、盈虧比、交易勝率等不進(jìn)行檢驗(yàn)。交易檢測結(jié)果所包含的時(shí)段也甚是有限,有過度數(shù)據(jù)發(fā)掘的可能,在此基礎(chǔ)上得出的測試結(jié)果,可信度無從判斷。 本文在考慮了以上問題基礎(chǔ)上,試圖設(shè)計(jì)一個(gè)具備穩(wěn)定盈利能力的程序化交易模型。首先從有效市場假說和行為金融學(xué)的相關(guān)理論出發(fā),建立了實(shí)現(xiàn)超額利潤的理論基礎(chǔ),根據(jù)投資者的非理性行為,市場定價(jià)過程中可能出現(xiàn)大幅度的單向價(jià)格運(yùn)動(dòng)。之后探討了程序化交易模型的運(yùn)行機(jī)制,對(duì)趨勢跟蹤這一交易模式做了全面的分析并利用R/S分析法驗(yàn)證了趨勢波動(dòng)的存在,趨勢波動(dòng)正是為交易模型提供利潤的核心。論文引入了自適應(yīng)通道來構(gòu)建具體的交易模型,自適應(yīng)通道在各類行情中有較好的泛用性,可以根據(jù)當(dāng)前的行情狀態(tài)進(jìn)行自主調(diào)整處理,在大趨勢中投資者的頭寸不至于由于短期回調(diào)而提前平倉,最大限度保證了盈利空間,而在橫盤期間形成強(qiáng)勢突破時(shí),可以保證快速進(jìn)場建立頭寸。最后,引入了大量品種的長期數(shù)據(jù)對(duì)模型的盈利能力和穩(wěn)定性進(jìn)行驗(yàn)證,結(jié)果表明自適應(yīng)通道突破模型具有較強(qiáng)的獲利能力,對(duì)投資者實(shí)盤交易具有一定的指導(dǎo)意義。
[Abstract]:Getting excess returns has always been the ultimate pursuit of the vast majority of investors involved in financial transactions, and effective and scientific trading depends on a set of effective trading strategies, whether fundamental or technical. In essence, all have their own internal strategies. Programmed trading is the process of quantifying the investment strategy. The designed strategy is identified and executed by the computer. In developed foreign markets, whether for individual investors or large institutional investors, Procedural trading is the most advanced, most scientific investment model. Programmed trading can seek non-random profit opportunities in complex and disordered price fluctuations. This kind of non-random profit opportunity is represented by the memory of the market price movement, that is, the price usually moves in one direction until the end of the stage, and once the direction is changed, The price movement will continue in the other direction. This memory feature is defined as a trend, and the existence of the trend provides a design direction for programmed transactions. At present, the article about programming transaction model design in our country is almost blank. Most of the strategies related to trading strategies are very simple, such as the ordinary moving average system, MACD,KDJ and other classic indicators derived from the trading system. Most of these tests verify the characteristics of positive returns, but do not test the degree of stability of the system capital curve, the withdrawal ratio of funds, the ratio of profit to loss, the ratio of trade success and so on. The period of time included in the results of transaction detection is also very limited, and there is the possibility of excessive data mining. On the basis of this, the reliability of the test results can not be judged. On the basis of considering the above problems, this paper attempts to design a program trading model with stable profitability. Based on the theory of efficient market hypothesis and behavioral finance, this paper establishes the theoretical basis of realizing excess profit. According to the irrational behavior of investors, there may be a large one-way price movement in the process of market pricing. Then it discusses the operating mechanism of the programmed trading model, makes a comprehensive analysis of the trading model of trend tracking and verifies the existence of the trend volatility by using the R / S method, which is the core of providing profit for the trading model. In this paper, adaptive channel is introduced to build a specific transaction model. Adaptive channel has good universality in all kinds of market, and it can be independently adjusted according to the current market state. In the general trend, investors' positions do not close early because of a short-term correction, which maximizes the margin of profit, while in the event of a strong breakthrough in the horizontal market period, it can guarantee the rapid entry into the market to establish positions. Finally, a large number of long-term data are introduced to verify the profitability and stability of the model. The results show that the adaptive channel breakthrough model has a strong profitability, and has a certain guiding significance for investors' real trading.
【學(xué)位授予單位】：中國海洋大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：F830.49

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 康劍南;;證券投資技術(shù)分析存在的問題及對(duì)策[J];商場現(xiàn)代化;2010年31期

2 戴潔,武康平;中國股票市場技術(shù)分析預(yù)測力的實(shí)證研究[J];數(shù)量經(jīng)濟(jì)技術(shù)經(jīng)濟(jì)研究;2002年04期

3 孫碧波;移動(dòng)平均線有用嗎?——基于上證指數(shù)的實(shí)證研究[J];數(shù)量經(jīng)濟(jì)技術(shù)經(jīng)濟(jì)研究;2005年02期

4 陳卓思,宋逢明;圖形技術(shù)分析的信息含量[J];數(shù)量經(jīng)濟(jì)技術(shù)經(jīng)濟(jì)研究;2005年09期

，

本文編號(hào)：2381298