公司治理指標在財務危機預測: 以美國上市公司為例;Corporate government indicators apply in financial distressed problem: taking US-listed company for example

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/72636

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/72636

Title:	公司治理指標在財務危機預測: 以美國上市公司為例;Corporate government indicators apply in financial distressed problem: taking US-listed company for example
Authors:	田馥慈;TIEN, FU-TZU
Contributors:	資訊工程學系
Keywords:	財務危機預測;公司治理指標;Financial distress problem;corporate governance indicators
Date:	2016-10-11
Issue Date:	2017-01-23 17:09:19 (UTC+8)
Publisher:	國立中央大學
Abstract:	財務危機預測問題一直以來都是重要且被廣泛討論的問題，包含分類器的使用及特徵篩選方法的選擇，都是全球金融界及研究學者關注的問題。我們過去已在台灣公司資料下證明出使用公司治理指標及財務指標當特徵會比單純用財務指標當特徵得到較好的預測結果。但卻發現美國公司方面的探討並不完整，因此我們想知道在美國公司資料下，將公司治理指標加入以財務指標為主的特徵集中，是否也可以改善預測準確率？　　在本研究中，利用z_score及block作為特徵，提出一個利用門檻值判斷危機公司的Rule-based預測模型。在美國公司資料下，能顯著提升原本只使用z_score的模型的準確率。同時，本論文也介紹如何手動收集美國公司的CGIs的方法。 ;Financial distress problem (FDP) has been important and widely studied topic. According different classifiers, feature selection methods even the ensemble learning have been discussed. The past research had proved CGIs can improve predict model in Taiwan firms dataset, so we want to know if CGIs could improve predict model in US firms dataset. We use z-score and block to be feature, and propose a new model to judge company by threshold. It can improve the accuracy of FDP in US dataset. We also introduce how to hand collect CGIs of US company.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	322	View/Open

社群 sharing

Loading...