本文關鍵詞： 銀行間債券市場 評級機構 發(fā)行人付費 投資人付費 排序Logistic模型　出處：《復旦大學》2014年碩士論文　論文類型：學位論文

【摘要】：2005年以來,隨著我國債券品種的不斷豐富,銀行間債券市場規(guī)模迅速擴大,從而帶動了我國評級行業(yè)的高速發(fā)展。信用等級的高低將直接影響到企業(yè)債券發(fā)行成功與否及發(fā)行成本,評級機構在債券市場中的影響力越來越大。然而,我國評級機構基本都是采用發(fā)行人付費模式,潛在的利益沖突可能造成評級虛高。2010年8月,中國銀行間市場交易商協(xié)會代表全體會員出資設立中債資信評估有限責任公司(簡稱“中債資信”),成為國內(nèi)首家且是唯一一家采用投資人付費模式的評級機構。成立至今,中債資信已為數(shù)百家企業(yè)進行評級或提供主動評級,并出具評級報告。對比中債資信的評級結果和其他發(fā)行人付費模式下的評級結果,前者要明顯低于后者1-2個級別,即評級虛高的現(xiàn)象很有可能存在。在此背景下,本文選取同時具有中債資信評級和其他發(fā)行人付費評級機構評級的企業(yè)作為樣本,通過描述性統(tǒng)計分析不同付費模式下評級結果的差異及差異的分布情況,并分析造成上述差異的可能原因。然后,利用排序選擇模型對兩種模式下的評級模型進行實證檢驗,挖掘不同模式下評級模型各自所關注的指標及各指標的影響程度,并比較兩者之間的差異。同時,運用多元線性回歸模型進一步分析影響兩種模式下級差大小的具體因素。此外,本文對國外付費模式的演變做了一定的梳理,其中既包括改進模式,如政府裁決模式、投資人所有評級機構模式、集中結算平臺模式和基于評級結果的付費模式等,也包括替代模式,如投資人付費模式和公用事業(yè)模式等,以便為推動我國評級機構付費模式改革提供借鑒。最后,本文簡要分析了投資人付費模式下可能存在的“搭便車”、“利益沖突”等問題,并針對上述問題,提出若干建議。
[Abstract]:Since 2005, with the increasing variety of bonds in China, the scale of interbank bond market has expanded rapidly. The credit rating will directly affect the success and cost of corporate bond issuance, and the rating agencies will have more and more influence in the bond market. Our country's rating agencies are basically using the issuer's payment model, the potential conflict of interest may lead to false ratings. August 2010, The China Interbank Market Dealers Association, on behalf of all its members, has contributed to the establishment of China National debt Credit Appraisal Co., Ltd. (hereinafter referred to as "China debt Credit"), becoming the first and only rating agency in China to adopt the investor-paying model. Since its establishment, China debt credit has rated hundreds of enterprises or offered active ratings, and issued rating reports. Compared with the rating results of the Chinese debt credit rating and other issuers' payment mode, the former is significantly lower than the latter by 1-2 grades. That is, the phenomenon of false ratings is likely to exist. In this context, this paper selects enterprises with credit rating of Chinese debt and other issuer rating agencies as samples. This paper analyzes the difference and distribution of rating results under different payment modes by descriptive statistics, and analyzes the possible causes of the above differences. Then, the ranking selection model is used to test the rating models under the two models. Mining the indexes concerned by different models and the degree of impact of each index, and comparing the differences between the two. At the same time, Using the multiple linear regression model to further analyze the specific factors that affect the difference between the two models. In addition, this paper makes a certain combing of the evolution of the foreign payment model, which includes the improved model, such as the government ruling model. The investor model of all rating agencies, the centralized settlement platform model and the payment model based on rating results, and so on, also include alternative models, such as investor payment model and public utility model, etc. Finally, this paper briefly analyzes the possible problems of "free rider", "conflict of interest" and so on under the investor-paid mode, and puts forward some suggestions in view of the above problems.
【學位授予單位】：復旦大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：F832.51

【參考文獻】

相關期刊論文 前3條

1 劉輝,周慧文;融合、共進與發(fā)展:資信評級行業(yè)中外分析研究[J];財經(jīng)論叢(浙江財經(jīng)學院學報);2003年01期

2 于立勇,詹捷輝;基于Logistic回歸分析的違約概率預測研究[J];財經(jīng)研究;2004年09期

3 岳振宇;;危機下的反思:失信的信用評級及其制度完善[J];證券市場導報;2009年09期

相關碩士學位論文 前1條

1 高民;信用評分模型的開發(fā)及probit回歸在模型中的應用[D];山東大學;2012年

，

本文編號：1515548