政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/122814

政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/122814

English | 正體中文 | 简体中文 | Post-Print筆數 : 27 | 全文笔数/总笔数 : 112721/143689 (78%)
造访人次 : 49530545 在线人数 : 823

RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

搜寻范围

查询小技巧：

您可在西文检索词汇前后加上"双引号"，以获取较精准的检索结果

若欲以作者姓名搜寻，建议至进阶搜寻限定作者字段，可获得较完整数据

进阶搜寻

主页 ‧ 登入 ‧ 上传 ‧ 说明 ‧ 关于政大典藏 ‧ 管理

到手机版

政大機構典藏 > 商學院 > 企業管理研究所（MBA學位學程） > 學位論文 > Item 140.119/122814

请使用永久网址来引用或连结此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/122814

题名:	台灣地雷股預警模型 Taiwan Financial Crisis Model
作者:	郭昭廷 Kuo, Chao-Ting
贡献者:	謝明華周冠男郭昭廷 Kuo, Chao-Ting
关键词:	企業授信財務危機預警模型邏輯斯回歸決策樹支持向量機機器學習 Python Scikit-Learn 特徵工程 Corporate loan Financial crisis model Logistic regression Decision tree Support vector machine Machine learning Python Scikit-Learn Feature engineering
日期:	2019
上传时间:	2019-04-01 15:12:31 (UTC+8)
摘要:	金融業對於國家經濟發展向來扮演重要角色，其中銀行業的企業授信的角色相當重要，若銀行擁有的是不良債權，將可能讓銀行面臨被倒帳的結果，進而可能對作為銀行一大部分的資金來源的一般或機構投資人產生負面影響，故若能有效建構財務危機預警模型，將能避免上述情形的發生。關於建構財務危機預警模型之文獻已汗牛充棟，本研究之差別在於以邏輯斯回歸、決策樹、支持向量機等三種機器學習方法建構模型，並以不進行抽樣，且不對財務比率進行歸納之方式，並運用Python程式之套件Scikit-Learn實作模型，最後加入另一個由銀行業界專家問卷進行特徵工程所獲得的模型和未加入問卷的模型進行比較，希望讓本研究建構之模型於預測效能上有不錯的表現。根據本研究實證發現，未加入問卷的模型的效能表現皆非常不錯，然而由銀行業界專家問卷進行特徵工程所獲得的模型的效能表現和前者相比下降居多，而兩者皆為了正確判別出財務危機資料而錯殺了不少的財務正常資料。 Financial Industry has been an issue on country’s economic growth. Among the financial industry, corporate loan has played an important role. If the loan the bank has is a non-performing loan, the bank could be faced with no payback, which could have a bad impact on personal or institutional investors. So, if we could build an effective financial crisis model, the condition mentioned above could be prevented from happening. There has been many research related to building financial crisis model. However, there are some differences between this one and such. First of all, we used three machine learning approaches, which include Logistic Regression, Decision Tree, Support Vector Machine, to build the models. Second, we did the research without sampling. Third, we didn’t make an induction to get specific variables for modeling. Fourth, we did modeling with Python’s Scikit-Learn package. And last, we designed a questionaire to get viewpoints from the professionals in the banking industry for us to do feature engineering to create another models, and compare the models with the ones without questionaire. We expect such difference could have good influence on the models’ performance. According to the result of empirical analysis, all of the models without questionaire have good performance. However, the performance of most of the models with questionaire have fallen when compared to the performance of the models without questionaire. And, no matter with or without questionaire, all of the models sacrificed crisis-free coporations to generate better performance on detecting corporations with crisis.
參考文獻:	[1] 行政院主計總處國民所得統計及國內經濟情勢展望. Available from: https://www.stat.gov.tw/public/data/dgbas03/bs4/ninews/10711/newtotal10711.pdf. [2] Beaver, W.H., Financial ratios as predictors of failure. Journal of accounting research, 1966: p. 71-111. [3] Altman, E.I., Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The journal of finance, 1968. 23(4): p. 589-609. [4] Ohlson, J.A., Financial ratios and the probabilistic prediction of bankruptcy. Journal of accounting research, 1980: p. 109-131. [5] Provost, F. and T. Fawcett, Data Science for Business: What you need to know about data mining and data-analytic thinking. 2013: " O`Reilly Media, Inc.". [6] Raschka, S., Python machine learning. 2015: Packt Publishing Ltd. [7] Zmijewski, M.E., Methodological issues related to the estimation of financial distress prediction models. Journal of Accounting research, 1984: p. 59-82. [8] 台灣經濟新報上市(櫃)公司基本資料資料庫. Available from: http://tejdb.tej.com.tw/ReportListing_new/reportlisting.aspx?UserCheck=3230313930313330313433373530544348494E4553452A&Report=wind. [9] Géron, A., Hands-on machine learning with Scikit-Learn and TensorFlow: concepts, tools, and techniques to build intelligent systems. 2017: " O`Reilly Media, Inc.". [10] 周志华, 机器学习. 2016: Qing hua da xue chu ban she. [11] Scikit-Learn choosing the right estimator. Available from: https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html. [12] Scikit-Learn Logistic Regression. Available from: https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html. [13] Scikit-Learn Decision Tree Classifier. Available from: https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier. [14] Scikit-Learn SVC. Available from: https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html#sklearn.svm.SVC. [15] 台灣經濟新報財務資料庫科目說明. Available from: http://tejdb.tej.com.tw/ReportListing_new/reportlisting.aspx?UserCheck=3230313930313330313430333331544348494E4553452A&Report=wm3. [16] TEJ信用風險觀測(TCRI)模組欄位說明. Available from: https://www.tej.com.tw/webtej/doc/crwatch1.htm.
描述:	碩士國立政治大學企業管理研究所(MBA學位學程) 105363083
資料來源:	http://thesis.lib.nccu.edu.tw/record/#G0105363083
数据类型:	thesis
DOI:	10.6814/THE.NCCU.MBA.018.2019.F08
显示于类别:	[企業管理研究所（MBA學位學程）] 學位論文

文件中的档案:

档案	大小	格式	浏览次数
308301.pdf	3222Kb	Adobe PDF2	2	检视/开启

在政大典藏中所有的数据项都受到原著作权保护.

社群 sharing

著作權政策宣告 Copyright Announcement

1.本網站之數位內容為國立政治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，惟仍請適度，合理使用本網站之內容，以尊重著作權人之權益。商業上之利用，則請先取得著作權人之授權。
The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

2.本網站之製作，已盡力防止侵害著作權人之權益，如仍發現本網站之數位內容有侵害著作權人權益情事者，請權利人通知本網站維護人員(nccur@nccu.edu.tw)，維護人員將立即採取移除該數位著作等補救措施。
NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.

DSpace Software Copyright © 2002-2004 MIT & Hewlett-Packard / Enhanced by NTU Library IR team Copyright © - 回馈