本文關(guān)鍵詞： 股價(jià)預(yù)測(cè) bp神經(jīng)網(wǎng)絡(luò) 粒子群算法 股票的可預(yù)測(cè)性　出處：《復(fù)旦大學(xué)》2013年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：隨著中國(guó)股票市場(chǎng)的發(fā)展,股票市場(chǎng)的投資活動(dòng)逐漸變得頻繁,股票市場(chǎng)逐漸成為證券市場(chǎng)中最活躍的市場(chǎng),股票成為投資者們最熱衷的投資產(chǎn)品。所以股票價(jià)格的預(yù)測(cè)成為了一項(xiàng)熱門(mén)研究。有效的預(yù)測(cè)分析方法可以很好的幫助投資者制定投資策略,在增加收益的同時(shí)降低風(fēng)險(xiǎn)。股票市場(chǎng)是一個(gè)非常復(fù)雜的系統(tǒng),但是它的內(nèi)在規(guī)律具有一定的趨勢(shì)性,而且受到經(jīng)濟(jì)政治等許多因素的影響,但就是如此股票市場(chǎng)這個(gè)系統(tǒng)的運(yùn)動(dòng)規(guī)律仍然是很難掌握的。許多的學(xué)者對(duì)股票市場(chǎng)進(jìn)行研究,并且產(chǎn)生了許多的方法模型。傳統(tǒng)的研究方法主要是基于數(shù)理統(tǒng)計(jì)理論的模型,先建立主觀(guān)的數(shù)據(jù)序列模型,再對(duì)模型進(jìn)行預(yù)測(cè)和研究,像時(shí)間序列等等模型在這方面有很多的應(yīng)用,但是其預(yù)測(cè)精度無(wú)法達(dá)到人們的要求,事實(shí)上人們?cè)谘芯恐兄饾u發(fā)現(xiàn)股票市場(chǎng)系統(tǒng)是一個(gè)復(fù)雜的非線(xiàn)性系統(tǒng),傳統(tǒng)的線(xiàn)性模型無(wú)法很好的逼近其內(nèi)在規(guī)律,許多的學(xué)者開(kāi)始研究股票的混沌性質(zhì),而且隨著非線(xiàn)性算法的發(fā)展,許多學(xué)者開(kāi)始使用神經(jīng)網(wǎng)絡(luò),遺傳算法等非線(xiàn)性算法對(duì)股票是場(chǎng)進(jìn)行預(yù)測(cè),各種基于非線(xiàn)性算法的股票預(yù)測(cè)模型被建立。 本文將粒子群優(yōu)化算法和bp神經(jīng)網(wǎng)絡(luò)進(jìn)行了融合,利用粒子群算法對(duì)神經(jīng)網(wǎng)絡(luò)的連接權(quán)重和閾值的訓(xùn)練進(jìn)行優(yōu)化,討論了各個(gè)參數(shù)的選取設(shè)定優(yōu)化,建立了粒子群優(yōu)化bp神經(jīng)網(wǎng)絡(luò)模型并將其用于股票預(yù)測(cè)的實(shí)證研究。通過(guò)對(duì)三種具有市場(chǎng)代表性的指數(shù)進(jìn)行實(shí)證分析,以及將粒子群優(yōu)化神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)效果與傳統(tǒng)的單一bp神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)效果進(jìn)行對(duì)比分析,得到的結(jié)果表明粒子群算法能夠有效的加強(qiáng)神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)能力,減小預(yù)測(cè)誤差,提高訓(xùn)練速度；三大市場(chǎng)的預(yù)測(cè)結(jié)果都比較理想,說(shuō)明了股票市場(chǎng)的可預(yù)測(cè)性；美國(guó)股票市場(chǎng)相對(duì)其他兩種市場(chǎng)具有更強(qiáng)的預(yù)測(cè)性,規(guī)律性更強(qiáng)。
[Abstract]:With the development of China's stock market, the investment activities of the stock market become more and more frequent, and the stock market gradually becomes the most active market in the stock market. Stocks have become the most popular investment products for investors. Therefore, the prediction of stock prices has become a hot research. Effective forecasting and analysis methods can help investors to make investment strategies. Stock market is a very complex system, but its inherent law has certain tendency, and is influenced by many factors such as economy and politics. However, it is still very difficult to master the movement law of the stock market system. Many scholars study the stock market. Traditional research methods are mainly based on mathematical statistics theory model. First, the subjective data sequence model is established, then the model is predicted and studied. Models such as time series have many applications in this field, but their prediction accuracy can not meet the requirements of people. In fact, people have gradually found that the stock market system is a complex nonlinear system. The traditional linear model can not approach its inherent law very well, many scholars begin to study the chaos property of stock, and with the development of nonlinear algorithm, many scholars begin to use neural network. The nonlinear algorithm such as genetic algorithm is used to predict the stock field, and a variety of stock prediction models based on nonlinear algorithm are established. In this paper, particle swarm optimization algorithm and BP neural network are fused, the training of connection weight and threshold of neural network is optimized by particle swarm optimization, and the selection and optimization of each parameter are discussed. The BP neural network model of particle swarm optimization is established and applied to the empirical research of stock forecasting. Through the empirical analysis of three representative indices of market. The prediction effect of PSO neural network is compared with that of traditional BP neural network. The results show that PSO can effectively enhance the prediction ability of neural network. Reduce the prediction error and improve the training speed; The forecast results of the three major markets are all satisfactory, which shows the predictability of the stock market; The US stock market is more predictable and more regular than the other two markets.
【學(xué)位授予單位】：復(fù)旦大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類(lèi)號(hào)】：TP18;F832.51

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 朱梅,王海燕;中國(guó)股票市場(chǎng)的非線(xiàn)性確定性預(yù)測(cè)[J];安徽工程科技學(xué)院學(xué)報(bào)(自然科學(xué)版);2004年02期

2 盧方元;經(jīng)濟(jì)預(yù)測(cè)的混沌分析[J];商業(yè)研究;2005年01期

3 馬明;李松;;基于遺傳算法優(yōu)化混沌神經(jīng)網(wǎng)絡(luò)的股票指數(shù)預(yù)測(cè)[J];商業(yè)研究;2010年11期

4 孫玉秋,陳圣滔;Bayes決策法在股票價(jià)格預(yù)測(cè)中的應(yīng)用[J];廣東技術(shù)師范學(xué)院學(xué)報(bào);2003年04期

5 姚洪興,盛昭瀚;股市預(yù)測(cè)中的小波神經(jīng)網(wǎng)絡(luò)方法的研究[J];管理工程學(xué)報(bào);2002年02期

6 王鳳蘭,聞邦椿;股價(jià)波動(dòng)序列的綜合預(yù)測(cè)法研究[J];經(jīng)濟(jì)經(jīng)緯;2005年02期

7 孟慶芳;彭玉華;;混沌時(shí)間序列改進(jìn)的加權(quán)一階局域預(yù)測(cè)法[J];計(jì)算機(jī)工程與應(yīng)用;2007年35期

8 李松;劉力軍;谷晨;;混沌時(shí)間序列預(yù)測(cè)模型的比較研究[J];計(jì)算機(jī)工程與應(yīng)用;2009年32期

9 梁夏;具有自糾錯(cuò)功能的人工神經(jīng)網(wǎng)絡(luò)在股票滾動(dòng)預(yù)測(cè)上的應(yīng)用[J];計(jì)算機(jī)應(yīng)用研究;1999年01期

10 陳輝煌;高巖;;證券市場(chǎng)的混沌現(xiàn)象分析[J];企業(yè)經(jīng)濟(jì);2009年06期

，

本文編號(hào)：1477704