Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/28028
標題: 台灣壽險業信用評等模型之研究
A Study on Credit Rating Model for Life Insurance Companies in Taiwan
作者: 吳秉楠
Wu, Bing-Nan
關鍵字: Credit Rating for Life Insurance companies;壽險業信用評等;Ordered Logit model;Back-Propagation Network model.;Ordered Logit模型;倒傳遞類神經網路模型
出版社: 應用經濟學系所
引用: 壹、中文部分 一、圖書 1.吳明隆(2006),「SPSS統計應用學習實務:問卷分析與應用統計」,知城數位,台北。 2.吳明隆、涂金堂(2006),「SPSS與統計應用分析」,五南圖書,台北。 3.李進生、謝文良、林允永、蔣炤坪、陳達新、盧陽正(2001),「風險管理-風險值(VaR)理論與應用」,清蔚科技,新竹。 4.林建智、王儷玲(2001),「美國保險業財務分析及清償能力追蹤系統之研究與建議」,保發中心,台北。 5.林師模、陳苑欽(2006),「多變量分析:管理上的應用」,雙葉書廊,台北。 6.周大慶、沈大白、張大成、敬永康、柯瓊鳳(2002),「風險管理新標竿-風險值理論與應用」,智勝文化,台北。 7.施才憲(2003),「財務報表分析」,東業企業,高雄。 8.陳耀茂、殷純淵(2004),「類神經網路 PCNeuron 使用手冊」,鼎茂圖書,台北。 9.陳遠哲、鄭純農、傅文芳(2003),「保險會計理論與實務」,保發中心,台北。 10.黃登源(1998),「應用迴歸分析」,華泰文化,台北。 11.黃俊英(2000),「多變量分析」,華泰文化,台北。 12.葉怡成(2004a),「類神經網路模式應用與實作(八版)」,儒林圖書,台北。 13.葉怡成(2004b),「應用類神經網路 (三版)」,儒林圖書,台北。 14.鄭濟世(1998),「我國壽險業資本適足性之研究」,保發中心,台北。 15.蔡政憲、彭金隆、許文彥、梁正德、洪炳輝、張士傑、莊聲和、喬治華、黃芳文(2007),「保險財務評估與監理」,保發中心,台北。 16.顏月珠(2001),「應用數理統計學」,三民書局,台北。 17.簡松棋(2005),「保險會計-原理與實務」,三民書局,台北。 二、期刊論文 1.丁玉成(2000),「臺灣區銀行信用評等之模式研究-以BankWatch評等為基礎的實證研究」,國立臺灣大學商學研究所博士論文。 2.王儷珊(2001),「我國產物保險公司清償能力的探討」,國立中山大學財務管理學系研究所碩士論文。 3.呂嘉盈(2000),「台灣產險產業保險財務研究-Logistic 模型之運用」,國立高雄第一科技大學保險營運系研究所碩士論文。 4.李美樺(2002),「台灣綜合證券商信用評等實證模型之研究」,國立中正大學企業管理研究所碩士論文。 5.施佳華(2001),「產險業信用評等模式之研究-美國產險公司之實證分析」,國立政治大學風險管理與保險學系研究所碩士論文。 6.施孟隆(1998),「農會信用部經營危機預警模式之研究」,國立中興大學農業經濟學系研究所博士論文。 7.施孟隆、蕭淑華、黃炳文、林三立(1999),「人壽保險公司評等系統之研究-因子分析法之運用」,壽險季刊,第113期,pp.43-68。 8.施孟隆、蕭淑華、黃炳文、林三立(2001),「台灣地區人壽保險公司評等系統之研究-多元Ordered Probit模型之應用」,保險專刊,第66輯,pp.27-47。 9.施麗玉(2002),「農會信用部財務危機預測模型之研究-模糊類神經網路系統之應用」,國立中興大學農業經濟學系研究所博士論文。 10.紀宗利(2003),「建構我國產險業信用評等制度之研究」,國立高雄第一科技大學風險管理與保險學系研究所碩士論文。 11.洪燦楠(2006),「台灣壽險業發展歷程與展望(上)」,壽險季刊,第142期,pp.7-19。 12.馬中驍(1996),「台灣地區壽險業清償能力預警模型-LOGIT與類神經網路之應用」,逢甲大學保險學研究所碩士論文。 13.高子荃、詹淑慧(2001),「壽險業喪失清償能力信賴區間之研究」,保險專刊,第63輯,pp.101-121。 14.郝充仁、周林毅(2003),「台灣地區人壽保險業經營績效因素分析之研究」,保險專刊,第19卷第1期,pp.75-105。 15.郭素陵(2002),「本國銀行信用評等實證模型之研究」,國立中正大學企業管理研究所碩士論文。 16.張大成、劉宛鑫、沈大白(2002),「信用評等模型之簡介」,中國商銀月刊,第21卷第11期,pp.1-5。 17.陳錦村、江玉娟、朱育男(2006),「商業銀行如何建置符合新巴塞爾資本協定的信用評等制度」,金融風險管理季刊,第2卷第1期,pp.115-140。 18.曾令寧、黃仁德(1997),「美國修正後的統一金融機構評等制度」,台灣經濟金融月刊,第33卷第4期,pp.18-25。 19.楊士昌(2002),「壽險業信用評等模式之研究-美國壽險公司之實證分析」,國立政治大學風險管理與保險學系研究所碩士論文。 20.蔡碩倉(1999),「台灣地區農會信用部金融預警評等系統之研究」,國立中興大學農業經濟學系研究所博士論文。 三、網路資源 1.中華信用評等公司,http://www.taiwanratings.com/tw/。 2.台灣經濟新報,http://www.tej.com.tw/。 3.行政院金融監督管理委員會,http://www.fscey.gov.tw/。 4.行政院金融監督管理委員會銀行局,http://www.banking.gov.tw/。 5.行政院金融監督管理委員會保險局,http://www.ib.gov.tw/。 6.行政院金融監督管理委員會檢查局,http://www.feb.gov.tw/。 7.保險事業發展中心,http://www.tii.org.tw/index.asp。 8.惠譽國際信用評等公司,http://www.fitchratings.com.tw/zh-tw/。 貳、西文部分 一、Journal Articles 1.Altman, E.I.(1968), “Financial Ratios, Discriminant Analysis and The Prediction of Corporate Bankruptcy”, Journal of Finance, 23(4):589-609. 2.Ambrose, J.M.,and Carroll, A.M.(1994), “Using Best''s Ratings in Life Insurer Insolvency Prediction”, Journal of Risk and Insurance, 61(2):317-327. 3.Ambrose, J.M.,and Seward, J.A.(1988), “Best''s Ratings, Financial Ratios and Prior Probabilities in Insolvency Prediction”, Journal of Risk and Insurance, 55(2):229-244. 4.Brockett, P.L., Cooper, W.W., Golden, L.L.,and Pitaktong, U.(1994),“A Neural Network Method for Obtaining an Early Warning of Insurer Insolvency”, Journal of Risk and Insurance, 61(3):402-424. 5.Barniv, R.,and Hershbarger, R.A.(1990),“Classifying Financial Distress in the Life Insurance Industry”, Journal of Risk and Insurance, 57(1):110-136. 6.Barniv, R.,and McDonald, J.B.(1992),“Identifying Financial Distress in the Insurance Industry : A Synthesis of Methodological and Empirical Issues”, Journal of Risk and Insurance, 59(4):543-574. 7.Cummins, J.D., Grace, M.F.,and Phillips, R.D.(1999), “Regulatory Solvency Prediction in Property-Liability Insurance : Risk-Based Capital, Audit Ratios, and Cash Flow Simulation”, Journal of Risk and Insurance, 66(3):417-458. 8.Carson, J.M.,and Hoyt, R.E.(1995),“Life Insurer Financial Distress: Classification Models and Empirical Evidence”, Journal of Risk and Insurance, 62(4):764-775. 9.Grace, M.F., Harrington, S.E.,and Klein, R.W.(1998), “Risk-Based Capital and Solvency Screening in Property-Liability Insurance : Hypotheses and Empirical Tests”, Journal of Risk and Insurance, 65(2):213-243. 10.Huang, C.S., Dorsey, R.E.,and Boose, M.A.(1994), “Life Insurer Financial Distress Prediction : A Neural Network Model”, Journal of Insurance Regulation, 13(2):131-167. 11.Lee, S.H.,and Urrutia, J.L.(1996), “Analysis and Prediction of Insolvency in the Property-Liability Insurance Industry : A Comparison of Logit and Hazardous Models”, Journal of Risk and Insurance, 63(1):121-130. 12.Pottier, S.W.,and Sommer, D.W.(1999), “Property-Liability Insurer Financial Strength Ratings : Differences Across Rating Agencies”, The Journal of Risk and Insurance, 66(4):621-642. 13.Trieschmann, J.S.,and Pinches, G.E.(1973), “A Multivariate Model for Predicting Financially Distressed Property-Liability Insurers”, Journal of Risk and Insurance, 40(3):327-338.
摘要: 
摘要
本研究利用公開的壽險業財務與非財務資料,運用因素分析法篩選重要變數建立信用評等構面,並進一步採用常態分配的觀念,建立適用於台灣且具有客觀性與科學性的壽險業信用評等模型。希望透過一個適當的分類與評等標準,將壽險業複雜的財務與業務狀況,用簡明的方式呈現,讓消費者與投資人等易於瞭解壽險公司的經營狀況,而研究結果可作為監理機關監理制度的參考指標。此外,亦透過Ordered Logit模型與倒傳遞類神經網路模型來建立信用評等預測模型,並比較其預測能力及可靠度。
經由實證分析可獲得下列幾點結論:
一、採用因素分析篩選重要變數所建立的信用評等構面分別為「資本適足性、資產品質及流動性(因素一)」、「獲利能力及管理績效(因素二)」、「獲利能力、管理績效、資產品質及資本適足性(因素三)」、「資本適足性及管理績效(因素四)」、「市場風險敏感性(因素五)」、「流動性(因素六)」等六大構面。
二、 稅前損益對總收入比率(E)、 資產總額週轉率(A)及 通貨膨脹率(S)在兩模型中是影響評等結果的共同變數指標,而指標屬性以獲利能力(E)的影響最為關鍵。
三、 有無加入金控(A)此變數對於評等結果並無顯著影響,而總體經濟變數方面,僅 通貨膨脹率(S)對於評等等級有顯著的影響。
四、當輸出變數為五個等級時,Ordered Logit模型原始樣本分類正確率(82.5%)高於倒傳遞類神經網路模型一及模型四之測試樣本(64.583%及60.417%)。當輸出變數為二元分類時,仍然以Ordered Logit模型總分類預測正確率較佳。而以等級E作為判斷壽險公司是否會被接管的臨界點,其分類預測正確率高於以等級D、E的情形。判斷可能被接管公司的預測正確率仍然以Ordered Logit模型較佳,而正常營運公司預測正確率的判斷,兩者差異不大。在型Ⅰ誤差方面,以Ordered Logit模型分類所得的結果較小;而以等級E作為判斷壽險公司是否會被接管的臨界點,其所得的型Ⅰ誤差高於以等級D、E作為臨界點的情形。
五、當評等等級較低時,可能發生違約風險而被監理機關接管的機率會較高,反之,則較低。

Abstract
Based on published financial and non-financial data from Life Insurance companies,Factor Analysis was adopted in this paper to select important variables to establish credit rating dimensions. In addition, the concept of normal distribution was used to establish objective and scientific credit rating model that matched the conditions of Life Insurance companies in Taiwan. Through proper categorization and evaluation standards,the complicated financial and business status of Life Insurance companies was expressed in a simple manner to let consumers, investors, and others understand the operation conditions of Life Insurance companies. Also the supervisory institution could take the results as the reference index for supervisory organization systems. Furthermore, Ordered Logit and the Back-Propagation Network models were adopted to establish predictive models for credit rating,and their predictive abilities and reliabilities were compared.
The empirical results were concluded as followings:
1. The credit rating dimensions established by Factor Analysis to select important variables were “captial adequacy, asset quality and liquidity (factor one)”, “earnings and management performance (factor two)”, “earnings, management performance, asset quality and captial adequacy (factor three)”, “captial adequacy and management performance (factor four)”, “sensitivity to market risks (factor five)”, “liquidity (factor six)”.
2.“ Income before taxes to Total Revenue(E)”, “ Total Assets Turnover Ratio(A)”, “ Inflation”were the common variables for Ordered Logit and the Back-Propagation Network models,and earnings was the most dominated one for the management indicator feature.
3. Join the Financial Holding Company did not affect the result of credit rating,and Inflation was the important variable that affect credit rating for Life Insurance companies.
4. When outputting parameters were five grades,the correct rate (82.5%) in primitive sample of Ordered Logit model was higher than test sample of the Back-Propagation Network model one and model four (64.583% and 60.417%) . When outputting parameters were two grades,the correct rate was better in Ordered Logit model. And we would regard grade E as the critical point whether Life Insurance companies would be taken over, the correct rate was higher than with the situation of grade D, E.The correct rate of Life Insurance companies would be taken over was still better in Ordered Logit model,and the judgement of the correct rate in the normal running company was higher for grade E and grade D, E .TypeⅠerror of Ordered Logit model was lower than others.And we would regard grade E as the critical point whether Life Insurance companies would be taken over, the result was higher than the situation of regarding grade D, E as the critical point.
5. When the grade of credit rating for Life Insurance companies was lower, they could take place default risk and the probability of Life Insurance companies would be taken over by the supervisory institution were relatively higher, on the contrary, it was relatively lower.
URI: http://hdl.handle.net/11455/28028
其他識別: U0005-2801200814292800
Appears in Collections:應用經濟學系

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.