本文關(guān)鍵詞：基于深度學(xué)習(xí)的股票價(jià)格趨勢(shì)預(yù)測(cè)方法研究　出處：《云南財(cái)經(jīng)大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 深度學(xué)習(xí) 受限的玻爾茲曼機(jī) 股票價(jià)格預(yù)測(cè) CD算法

【摘要】：當(dāng)今股票市場(chǎng)不僅為優(yōu)秀掛牌企業(yè)提供融資,同時(shí)讓一些有投資意識(shí)的股民提供資金出路。從而使得社會(huì)資源得到更好的配置和宏觀經(jīng)濟(jì)得以調(diào)控,然而由于股市的不確定性,每個(gè)投資人對(duì)股市認(rèn)知的異同性、技術(shù)分析的復(fù)雜性等因素,使得廣大股民投資的回報(bào)率達(dá)不到預(yù)期,有的甚至血本無(wú)歸。所以一直以來(lái)股票市場(chǎng)都被無(wú)論是政府、企業(yè)還是投資者所高度關(guān)注。股票價(jià)格趨勢(shì)的預(yù)測(cè)更是股票研究中的熱點(diǎn)。眾所周知,由于股市的波動(dòng)具有極強(qiáng)的非線性、高噪聲等特點(diǎn),所以對(duì)股票價(jià)格趨勢(shì)預(yù)測(cè)極其困難,傳統(tǒng)股票預(yù)測(cè)方法往往收效甚微。因此如何建立新的股票價(jià)格趨勢(shì)預(yù)測(cè)的模型來(lái)提高預(yù)測(cè)的準(zhǔn)確度,從而幫助金融投資者有效規(guī)避風(fēng)險(xiǎn),投資獲利最大化,具有重要的理論意義和應(yīng)用價(jià)值。本文首先闡述了傳統(tǒng)股票預(yù)測(cè)方法,大體分為:基本分析法主要是從宏觀微觀經(jīng)濟(jì)、相關(guān)公司的財(cái)務(wù)報(bào)表和現(xiàn)金流等信息角度,通過(guò)相對(duì)估值和折現(xiàn)估值等等,對(duì)該股票的內(nèi)在價(jià)值進(jìn)行估值。不足之處:信息不對(duì)等性及相關(guān)掛牌公司披露信息延時(shí)性、準(zhǔn)確性等導(dǎo)致估值困難。大盤分析法主要是依據(jù)統(tǒng)計(jì)圖表,如K線圖,其形態(tài)可分為整理形態(tài)和趨向線等,根據(jù)對(duì)其特定的形態(tài)來(lái)判斷股市的未來(lái)動(dòng)向。不足之處:此類分析方法繁多,且各個(gè)投資人判斷習(xí)慣不同,方法之間存在巨大差別。統(tǒng)計(jì)學(xué)分析法主要是采用最小二乘構(gòu)建各種回歸,例如混合回歸模型、自回歸模型等進(jìn)行股票價(jià)格趨勢(shì)預(yù)測(cè),此類模型的預(yù)測(cè)預(yù)測(cè)準(zhǔn)確率較前兩類預(yù)測(cè)方法要高。不足之處:這些回歸模型通常假設(shè)前提太多,且對(duì)非線性強(qiáng)的問(wèn)題處理能力,而股票價(jià)格趨勢(shì)的預(yù)測(cè)問(wèn)題影響因素眾多且非線性極強(qiáng)�；谌斯ど窠�(jīng)網(wǎng)絡(luò)的預(yù)測(cè)模型具有高度自組織、自調(diào)整和自學(xué)習(xí)的能力、是一個(gè)復(fù)雜度極高的非線性系統(tǒng),模型預(yù)測(cè)結(jié)果通常也要優(yōu)于上述傳統(tǒng)方法。不足之處:基于神經(jīng)網(wǎng)絡(luò)的股票預(yù)測(cè)模型容易陷入局部最小值的問(wèn)題,且多層神經(jīng)網(wǎng)絡(luò)在對(duì)復(fù)雜事物的描述時(shí),往往要增多隱含層的層數(shù),這樣會(huì)導(dǎo)致梯度擴(kuò)散的問(wèn)題,從而影響準(zhǔn)確率。本文正是從基于人工神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型的缺點(diǎn),如梯度擴(kuò)散和局部最小值等問(wèn)題,從而提出了采用受限的玻爾茲曼機(jī)模型構(gòu)建基于深度學(xué)習(xí)的股票價(jià)格趨勢(shì)預(yù)測(cè)模型。深度學(xué)習(xí)是基于神經(jīng)網(wǎng)絡(luò)基礎(chǔ)上發(fā)展而來(lái),不僅繼承了神經(jīng)網(wǎng)絡(luò)方法的優(yōu)點(diǎn),而且很好的克服了神經(jīng)網(wǎng)絡(luò)方法的不足之處。本文預(yù)測(cè)模型采用受限的玻爾茲曼機(jī)來(lái)構(gòu)建深度置信網(wǎng)絡(luò),學(xué)習(xí)方法是采用K步吉布斯采樣后,結(jié)合對(duì)比散度算法,來(lái)訓(xùn)練整個(gè)深度置信網(wǎng)絡(luò)。最后利用收集的格力空調(diào)的股票價(jià)格信息來(lái)訓(xùn)練本文預(yù)測(cè)模型并對(duì)本文模型預(yù)測(cè)準(zhǔn)確率進(jìn)行了檢驗(yàn)。選用基于BP神經(jīng)網(wǎng)絡(luò)的股票價(jià)格預(yù)測(cè)模型作為本文預(yù)測(cè)模型的對(duì)比模型,并用采用實(shí)例貴州茅臺(tái)和比亞迪的股票價(jià)格信息來(lái)檢驗(yàn)兩個(gè)模型的預(yù)測(cè)準(zhǔn)確率,實(shí)驗(yàn)結(jié)果表明:基于深度學(xué)習(xí)的股票價(jià)格趨勢(shì)預(yù)測(cè)模型效果良好,且準(zhǔn)確率要優(yōu)于BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型。本文創(chuàng)新點(diǎn):(1)本文采用了基于受限的玻爾茲曼機(jī)構(gòu)建深度置信網(wǎng)絡(luò)的股票價(jià)格趨勢(shì)預(yù)測(cè)模型,學(xué)習(xí)方法采用了經(jīng)過(guò)K步的吉布斯采樣后的對(duì)比散度算法(CD算法)來(lái)訓(xùn)練預(yù)測(cè)模型。最后給出實(shí)例驗(yàn)證。(2)將本文預(yù)測(cè)模型與基于BP神經(jīng)網(wǎng)絡(luò)股票價(jià)格趨勢(shì)預(yù)測(cè)模型的預(yù)測(cè)準(zhǔn)確率進(jìn)行了實(shí)例比較。
[Abstract]:The stock market not only provides financing for outstanding listed companies, while some investment minded investors to provide funds to make way. A better allocation of social resources and macroeconomic regulation to, however due to the uncertainty in the stock market, stock market investors on the similarities and differences of each cognitive complexity, factors such as technical analysis, making the stock investment the rate of return is not up to expectations, some even lose everything. So since the stock market has been highly concerned by both the government, enterprises and investors. The stock price trend forecast is a hot stock research. As everyone knows, because of the volatility of the stock market has a very strong nonlinear, high noise characteristics, so the stock price trend prediction is extremely difficult, the traditional stock forecasting methods often have little effect. So how to establish a new prediction model of stock price trend To improve the prediction accuracy, so as to help investors to avoid financial risk effectively, investment profit maximization, and has important theoretical significance and application value. This paper describes the traditional prediction methods of stock, can be divided into: basic analysis mainly from the macro and micro economy, financial statements and cash flow information related to the company's point of view. Through the relative valuation and valuation discount and so on, the intrinsic value of the stock valuation. Disadvantages: information asymmetry and listed company information disclosure delay, as a result of the valuation accuracy difficult. Large disk analysis method is mainly based on the statistical charts, such as the K map, its shape can be divided into the consolidation pattern and the trend line. According to the future trends of its specific form to determine the stock market. The shortcomings of such analysis methods are various, and each investor to judge different habits, there is a huge difference between the methods. The statistical analysis method is mainly constructed by least squares regression, such as mixed regression model, auto regression model to forecast the stock price trend forecast of this kind of model accuracy than prediction method to high. The first two shortcomings: the regression model is usually premise too much, and the ability to deal with the nonlinear problems. The factors affecting the prediction of stock price trend of large and highly nonlinear. The prediction model based on artificial neural network has a high degree of self-organization, self adjustment and self-learning ability, is a very complex nonlinear system, the prediction results are better than the traditional method. Disadvantages: neural network stock forecasting the model is easy to fall into the local minimum problem based on multilayer neural network and the complicated description of things, tend to increase the number of hidden layer, it will Lead to gradient diffusion problems, thus affecting the accuracy. This article is from the disadvantages of prediction model based on artificial neural network, such as the gradient diffusion and local minimum problem, and put forward the construction of deep learning of stock price trend forecast model based on the Boltzmann machine model is limited. Deep learning is developed based on neural network based on not only inherits the advantages of neural network method, and good overcomes the defects of the neural network method. In this paper, a prediction model based on Boltzmann machine limited to construct the deep belief network learning method is the use of K step of Gibbs sampling, comparing with the divergence algorithm, to train the entire depth of belief network. Finally the collection the GREE stock price information to train the prediction model and the prediction accuracy of this model is tested based on BP by God. The stock price prediction model as the prediction model comparison model, and by using examples Kweichow Moutai and BYD's stock price information to test the prediction accuracy of the model is two, the experimental results show that the deep learning of stock price trend forecast model based on good effect, and the accuracy is better than BP neural network prediction model. The innovation of this paper: (1) the prediction model of Boltzmann limited mechanism built deep belief networks of stock price trend based on learning method is adopted after comparing K step of Gibbs sampling after the divergence algorithm (CD algorithm) to train the prediction model. Finally, an instance is given. (2) the prediction model with the prediction of BP neural network prediction model of stock price trend based on the accuracy of the comparison.

【學(xué)位授予單位】：云南財(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;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張承釗;潘和平;;基于前向滾動(dòng)EMD技術(shù)的預(yù)測(cè)模型[J];技術(shù)經(jīng)濟(jì);2015年05期

2 戴淑庚;許俊;;我國(guó)股票市場(chǎng)財(cái)富效應(yīng)研究——基于消費(fèi)者信心與廣義虛擬經(jīng)濟(jì)的視角[J];廣義虛擬經(jīng)濟(jì)研究;2015年01期

3 劉慧;王聰;;我國(guó)城鎮(zhèn)居民股票資產(chǎn)財(cái)富效應(yīng)影響因素分析[J];金融與經(jīng)濟(jì);2015年01期

4 史美景;宋婷;;股票市場(chǎng)長(zhǎng)期波動(dòng)趨勢(shì)度量及影響因素分析——基于Spline-GARCH模型[J];數(shù)理統(tǒng)計(jì)與管理;2015年01期

5 孫志軍;薛磊;許陽(yáng)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

6 李成剛;田益祥;何繼銳;;AC算法的EMD分解GMDH組合的預(yù)測(cè)模型及應(yīng)用[J];系統(tǒng)管理學(xué)報(bào);2012年01期

7 楊海深;傅紅卓;;基于貝葉斯正則化BP神經(jīng)網(wǎng)絡(luò)的股票指數(shù)預(yù)測(cè)[J];科學(xué)技術(shù)與工程;2009年12期

8 崔建福,李興緒;股票價(jià)格預(yù)測(cè):GARCH模型與BP神經(jīng)網(wǎng)絡(luò)模型的比較[J];統(tǒng)計(jì)與決策;2004年06期

9 王上飛,周佩玲,吳耿峰,付忠謙,儲(chǔ)閱春,沈謙;徑向基神經(jīng)網(wǎng)絡(luò)在股市預(yù)測(cè)中的應(yīng)用[J];預(yù)測(cè);1998年06期

10 張健,陳勇,夏罡,何永保;人工神經(jīng)網(wǎng)絡(luò)之股票預(yù)測(cè)[J];計(jì)算機(jī)工程;1997年02期

相關(guān)會(huì)議論文 前1條

1 萬(wàn)杰;胡清華;劉金福;蘇鵬宇;于達(dá)仁;;基于深度學(xué)習(xí)理論的短期風(fēng)速多步預(yù)測(cè)方法研究[A];智能化電站技術(shù)發(fā)展研討暨電站自動(dòng)化2013年會(huì)論文集[C];2013年

相關(guān)碩士學(xué)位論文 前1條

1 翟曼;基于PSO優(yōu)化混沌BP神經(jīng)網(wǎng)絡(luò)的股票指數(shù)預(yù)測(cè)模型研究[D];河北大學(xué);2011年

，

本文編號(hào)：1377481