摘要近年来国内股票市场整体呈现低迷态势，这使得以追求绝对收益为目标的量 化交易得到了广泛的关注，现已成为全球基金行业的主要交易方法。因此，针对 股票市场的量化交易算法的研究有一定的理论和现实意义。本文阐述了量化交易 的基本概念、交易策略和交易模型，给出了支持向量机（SVM）的简要介绍。并利 用 SVM 的回归技术，以近 1 年沪深 A 股中流通性较好的银行业股票、证券业股票、 保险业股票的日线数据为对象，结合 EMA、MACD、KDJ 等指标，建立了 SVM 的股票 价格趋势预测模型。初步的测试结果表明了该算法在预测沪深 A 股市场流通性较 好的行业股票中有一定的有效性。68515

毕业论文关键词 量化交易 支持向量机 预测模型 回归分析

Title      Research of Quantitative Trading Algorithm  Based on SVM

Abstract

In recent years,China stock market has been in the doldrums.This makes quantitative investment which expects absolute returns get wide attention.So far,it has become a main investment approach of global fund industry.Therefore,it is of great theoretical and realistic significance to study quantitative methods in the stock market.This paper illustrates the basic concept,strategies and models of quantitative trading.And it also gives a brief introduction of Support Vector Machines(SVM).The paper uses the recent one year ’ s daily stock figures of the bank industry,securities industry and insurance industry in the markets of Shanghai and Shenzhen A share,which have good liquidity.It combines the EMA,MACD,KDJ and some other indices with regression techniques to establish a stock price prediction model.The preliminary results show that the algorithm has some validity in predicting the stocks which have better liquidity in the stock markets of Shanghai and Shenzhen A share.

Keywords Quantitative Trading SVM Prediction Model Regression Analysis

目 次

1 绪论 1

1.1 课题背景 1

1.2 国内外研究现状 2

1.3 课题主要研究内容 3

2 量化交易概述 4

2.1 量化交易的发展 4

2.2 量化交易分类 4

2.3 量化交易策略 5

3 统计学习理论与支持向量机 8

3.1 引言 8

3.2 统计学习理论简述 8

3.3 结构风险最小化原则 10

3.4 支持向量机 11

3.5 本章小结 13

4 基于 SVM 金融时间序列建模