摘要 市场商品价格的波动对企业及个人的决策往往有着重大的影响，尤以当前股票市场影响为最。股市价格的涨跌已然已然成为现代社会的“晴雨表”和“报警器”，在社会经济中有着举足轻重的地位，股票预测也越来越成为社会研究的重要对象。本文介绍了股票在中国的研究现状及现代股票预测的一些基本方法，首先以股市的价格均线为研究对象，试图通过控制理论的方法，拟合股市曲线，建立三阶数学模型，探讨股市均线波动的影响因素，并预测未来股市均线的波动趋势。但因股市显著的非线性特征，拟合函数产生了较大的误差，在进行相应的误差分析的基础上，本文又采用神经网络中的BP神经网络算法，训练拟合股价均线，最终得到了较为满意的结果。  87420

毕业论文关键词  股票预测 曲线拟合 最小二乘法 神经网络 BP算法

毕业设计外文摘要

Title   Analysis of zero point distribution of 3 order fitting   Model according to the market price trend curve of a product                                               

Abstract The fluctuation of market commodity prices has a great influence on the decision making of enterprises and inpiduals, especially in the current stock market。 The price rise of the stock market has already become the "barometer" of the modern society and the "alarm", which plays an important role in the social economy。 Is introduced in this paper the prediction of the research situation in China and the modern stock shares in some basic methods, to stock price moving average as the research object, trying to through the control theory method, fitting curve of the stock market, the establishment of third order mathematical model, to investigate the factors that influence the average volatility of the stock market, and forecasts the trend of fluctuations in the future of the stock market average。 However, due to the significant nonlinear characteristics of the stock market, the specific function of the fitting can not be achieved, this paper also uses the neural network BP neural network algorithm, training to fit the stock price moving average, and to seek short-term stock price movements。

Keywords  Stock forecast curve fitting least square method neural network          BP algorithm

目 次

1  绪论 1源-于Y优+尔-论.文:网www.chuibin.com 原文+QQ7520^18766

1。1  论文研究背景及意义 1

1。2  国内外研究现状 1

1。3  论文的研究方法及框架结构 4

2  中国股票市场的特征分析 5

2。1  股票价格波动的因素分析 5

2。2  股票预测的常用术语 7

3  曲线拟合及股票预测 9

3。1  问题描述 9

3。2  曲线拟合 9

3。3  模型建立 10

3。4  误差产生的原因分析 13

4  神经网络 15

4。1  神经网络的基本原理 15

4。2  BP神经网络及其算法 16

4。3  基于BP网络的预测模型