English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27533/39387
Visitors : 2540861      Online Users : 30
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/18173

Title: 基於財務能力分群的財務變數挑選流程:以中國和台灣上市櫃公司為例
An Integrated Approach to Financial Feature Selection Based on the Financial Ability Clustering : Two Case studies on Financial Distress Prediction Problem for Chinese and Taiwan Listed Companines
Authors: Ming-Hsun Tasi
蔡明勳
Contributors: NTOU:Department of Computer Science and Engineering
國立臺灣海洋大學:資訊工程學系
Keywords: 企業財務預測;變數挑選;支持向量機;財務能力分群
Financial Prediction;Feature Selection;SVM;Financial Ability Clustering
Date: 2010
Issue Date: 2011-07-04
Abstract: 近年來由於金融海嘯,造成了許多公司企業倒閉,為了要避免投資大眾的損失如何根據財務報表分析來預測危機公司的發生一直以來都是一門熱門的研究課題。 當Altman 開始以多變量分析建立危機模型後,近年來大部分都以探討新的建模演算以及使用非財務變數來預測危機公司,對於挑選變數組合顯少著墨。相較於近代越來越多的指標被提出,過去常用的挑選方法已經不適用於現今。本論文先探討過去財務界常用之挑選方法,並且提出他們之缺點主要有以下兩點,一有可能會挑出一些意義上高相關的財務指標,二有可能會挑出對於分類器而言無效的財務指標,我們根據Machine learning的挑選變數方法來解決這些的缺點。再根據財務報表分析將財務指標分群並且根據Machine learning挑選變數的方法用於挑選財務指標,建構出本論文財務指標挑選之流程,並且證明了此流程能夠比過去的人工挑選方法,或是統計過去文獻使用變數再使用統計挑選方法等等常見的方法來的有效,並且能夠適用於各種國家的財務指標挑選。 本論文主要有兩個貢獻。一: 我們提出一套將財務指標分群後挑選之流程,能夠比過去不分群的常用方法來的好,並且能夠解決過去方法之缺點。二: 我們利用中國以及台灣的危機預測證明了此套流程能夠適用於各種國家的財務指標挑選。
As the financial crisis in the past decades, resulting in closure of many companies, how to predict the financial statements of the occurrence of the crisis the company in order to avoid public investing loss has always been a popular research topic. After Altman E. I. starting use MDA model to predict business crisis, recent years most of researchers were using different Algorithm or using non-financial feature to predict, but feature selection method in financial research is very few. And recent years more new financial indexs are created, we think that the past feature selection method is not suitable for today’s feature selection . This thesis first discuss the feature selection method used in financial reasearch, then present their shortcomings, and provide the solution by using machine learning’s feature selection method. And we use financial analysis to clustering the financial index, according to the machine learning’s feature selection that Proposed an integrating financial feature selection is more effective then according to experts picking feature sets or using ANOVA,Logistic Regression,Discriminant Analysis to pick features . There are two major contributions. In this thesis, one of these is that develop a integrating financial features clustering selection method is more effective than other common feature selection methods in Chinese Business Prediction.Another contribution is that, proving this financial feature selection method is also can be used in every country by using China and Taiwan Business Prediction to proof .
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M97570006
http://ntour.ntou.edu.tw/ir/handle/987654321/18173
Appears in Collections:[資訊工程學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML153View/Open


All items in NTOUR are protected by copyright, with all rights reserved.

 


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback