【摘要】：股票市場(chǎng)在整個(gè)金融市場(chǎng)中起到很重要的作用,如何在股市中獲取有效的交易信號(hào)是股市投資一直在探討的話題。本文首先綜述了深度強(qiáng)化學(xué)習(xí)理論及模型,進(jìn)而以深度學(xué)習(xí)和強(qiáng)化學(xué)習(xí)為基礎(chǔ),結(jié)合深度強(qiáng)化學(xué)習(xí)相關(guān)理論模型,從自動(dòng)化股市投資交易決策機(jī)制構(gòu)建角度構(gòu)造股市深度強(qiáng)化學(xué)習(xí)模型。在股市投資策略中使用深度強(qiáng)化學(xué)習(xí)模型進(jìn)行策略的構(gòu)建是有效的,從各項(xiàng)策略評(píng)估指標(biāo)結(jié)果顯示深度強(qiáng)化學(xué)習(xí)模型對(duì)KD指標(biāo)交易信號(hào)抓取的有效性要比單純的KD指標(biāo)交易要有效,通過深度強(qiáng)化學(xué)習(xí)構(gòu)建的交易模型可以應(yīng)用到投資策略的構(gòu)建中。同時(shí)對(duì)個(gè)股的評(píng)估中發(fā)現(xiàn)深度強(qiáng)化學(xué)習(xí)策略是大概率獲利策略,需要分散投資來減少投資風(fēng)險(xiǎn),實(shí)現(xiàn)大概率獲利。本文構(gòu)建了以深度強(qiáng)化學(xué)習(xí)為理論基礎(chǔ)的股市投資策略模型,并通過實(shí)證數(shù)據(jù)驗(yàn)證了該模型的有效性,揭示了深度強(qiáng)化學(xué)習(xí)在股市投資策略構(gòu)建的內(nèi)在邏輯。這對(duì)投資者自動(dòng)化投資模型構(gòu)建、股市投資策略的構(gòu)建、人工智能在金融投資領(lǐng)域的應(yīng)用和提高投資者策略收益率都做出了有益的借鑒。
[Abstract]:Stock market plays an important role in the whole financial market. How to obtain effective trading signals in stock market is a topic that stock market investment has been discussing all the time. This paper first summarizes the theory and model of deep reinforcement learning, and then combines the related theory model of depth reinforcement learning with depth learning and reinforcement learning as the foundation. From the perspective of automatic stock market investment and trading decision-making mechanism, the paper constructs a stock market depth reinforcement learning model. It is effective to use the deep reinforcement learning model to construct the strategy in the stock market investment strategy. The results of each strategy evaluation index show that the depth reinforcement learning model is more effective than the pure KD index in grasping the trading signal of KD index. The transaction model constructed by deep reinforcement learning can be applied to the construction of investment strategy. At the same time, it is found in the evaluation of individual stocks that the deep reinforcement learning strategy is a high-probability profit-making strategy, which requires diversification to reduce the investment risk and realize the high-probability profit-making. In this paper, a stock market investment strategy model based on deep reinforcement learning is constructed, and the validity of the model is verified by empirical data, which reveals the inherent logic of deep reinforcement learning in the construction of stock market investment strategy. It can be used for reference for the construction of investor's automatic investment model, the construction of stock market investment strategy, the application of artificial intelligence in the field of financial investment and the improvement of the return rate of investor strategy.
【學(xué)位授予單位】：廣東財(cái)經(jīng)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：F832.51

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 余凱;賈磊;陳雨強(qiáng);徐偉;;深度學(xué)習(xí)的昨天、今天和明天[J];計(jì)算機(jī)研究與發(fā)展;2013年09期

2 孫志軍;薛磊;許陽明;王正;;深度學(xué)習(xí)研究綜述[J];計(jì)算機(jī)應(yīng)用研究;2012年08期

3 范波,潘泉,張洪才;一種基于分布式強(qiáng)化學(xué)習(xí)的多智能體協(xié)調(diào)方法[J];計(jì)算機(jī)仿真;2005年06期

4 高陽,周志華,何佳洲,陳世福;基于Markov對(duì)策的多Agent強(qiáng)化學(xué)習(xí)模型及算法研究[J];計(jì)算機(jī)研究與發(fā)展;2000年03期

5 張汝波,周寧,顧國昌,張國印;基于強(qiáng)化學(xué)習(xí)的智能機(jī)器人避碰方法研究[J];機(jī)器人;1999年03期

，

本文編號(hào)：2244990