毕业设计-时间序列分析模型研究.docVIP

- PAGE I - 本科毕业设计 （20 届） 时间序列分析模型研究 所在学院 专业班级 信息与计算科学 学生姓名 学号 指导教师 职称 完成日期 年 月 PAGE I 摘　要 【摘要】股价数据具有庞杂性、波动复杂性等等特点，造成了分析非常困难。对其进行时间序列建模是现代计量经济学最常用的手段。股市系统中时间序列的预测问题又具有重要的理论及实际意义。时间序列的获取是通过对数据库中数据进行分类汇总分析而获得。获取时间序列数据以后可以对它进行预测分析，从而较准确地预见股票价格的演进。文中介绍了时间序列的基本知识，同时比较了ARMA和GARCH两种常用模型，得出对于中国股市，GARCH模型性能优于ARMA模型。 【关键词】时间序列；ARMA模型；GARCH模型。 【ABSTRACT】Share data has the heterogeneous, volatility, and the complexity of the characteristics,which make the analysis result very difficult.Time-series econometric model is the most commonly used modern means. Market system for the time series prediction also has important theoretical and practical significance. Time series database access is through the pooled analysis of data obtained classification. Getting time-series data can later be analyzed to predict it, which more accurately predicted the evolution of share prices. This paper introduces the basics of time series, ARMA and GARCH also compared two commonly used models, obtained for the Chinese stock market, GARCH model is better than ARMA model. 【KEYWORDS】Time-series；ARMA model；GARCH model。 目　录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc292057888 摘　要 PAGEREF _Toc292057888 \h II HYPERLINK \l _Toc292057889 Abstract PAGEREF _Toc292057889 \h III HYPERLINK \l _Toc292057890 目　录 PAGEREF _Toc292057890 \h IV HYPERLINK \l _Toc292057891 1 绪论 PAGEREF _Toc292057891 \h 1 HYPERLINK \l _Toc292057892 1.1 引言 PAGEREF _Toc292057892 \h 1 HYPERLINK \l _Toc292057893 1.1.1 国内外研究现状 PAGEREF _Toc292057893 \h 1 HYPERLINK \l _Toc292057894 1.2 ARMA模型介绍 PAGEREF _Toc292057894 \h 2 HYPERLINK \l _Toc292057895 1.2.1 AR（p）模型 PAGEREF _Toc292057895 \h 2 HYPERLINK \l _Toc292057896 1.2.2 MA（q）模型 PAGEREF _Toc292057896 \h 3 HYPERLINK \l _Toc292057897 1.2.3 ARMA（p，q）模型 PAGEREF _Toc292057897 \h 3 HYPERLINK \l _Toc292057898 1.2.4 ARMA建模过程 PAGEREF _Toc292057898 \h 4 HYPERLINK \l _Toc29205789