【作者】 李大锋；

【导师】 罗林开；

【作者基本信息】 厦门大学 ， 模式识别与智能系统， 2008， 硕士

【摘要】 随着我国新基金的不断设立和基金行业的超常规发展,恰当的监测、分析和评价基金的业绩已越来越重要。在此背景下,本文希望通过借鉴西方国家证券投资基金绩效评价的方法,探索符合我国证券市场特点的业绩评估体系,为基金投资者、监管者、基金管理公司以及基金业的发展提供一些参考。统计学习理论是一种专门研究有限样本情况下机器学习规律的理论,追求在有限信息的条件下得到最优结果。支持向量机是在统计学习理论的基础上发展而来的一种新的机器学习方法,在解决有限样本、非线性及高维模式识别问题中表现出许多特有的优势。本文首先从理论上对国内外各种基金业绩评估的方法进行了较为系统的回顾和总结。其次,通过参考国内外不同基金评估所选用的指标,结合我国基金业的实际情况,分别从基金公司、基金经理、风险水平及风险调整前后的收益水平等多个方面,给出了基金评估的指标集;对于指标集的特征选择,主要采用独立成分分析方法和核主成分分析方法。然后,利用支持向量机模型,对我国开放式基金的业绩进行了实证评估。实证结果表明,在我国基金业绩评估中,本文的方法取得了较好的分类准确率,说明了本文方法的可行性及有效性。更多还原

【Abstract】 With the continuous establishment of new funds and the super development of funds industry in our country, properly inspecting, analyzing and evaluating funds performance is becoming more and more important. Under the background, the paper tries to discuss performance evaluation system of Chinese security investment funds. On the basis of the ways of performance evaluation in the developed countries, we hope to sump up some characteristics of Chinese funds and give some useful references to the relevant investors, asset management companies and the development of funds industry.Statistical learning theory is a theory of machine learning law dealing with small samples, and it takes into account the requirement of the generalization ability and the most excellent answer in limited conditions. Based on Statistical Learning Theory, a new machine learning method—support vector machine is put forward, and there are some virtues in dealing with the problem of pattern recognition, such as the problems of small samples, high dimensionality, non linearity.This paper gives an integrative introduction to correlative theories and models of funds performance evaluation in and abroad. On a basis of the domestic data sets of funds performance, the variables_analyzed by way of using the statistical identifying method._In this paper using the effective data mining algorithm—SVM classifier as a modeling method, and introduce Independent Component Analysis as a feature selection tool to effective select the better ratios of correlative indicators from funds data for the establishment of funds performance evaluation, thereby optimizing and improving the classification model performance based on support vector machines. By domestic funds data empirical analysis and comparison with other methods results, confirmed the validity and practicality of the funds performance evaluation model through independent component analysis and support vector machine classifier established.更多还原