本文關(guān)鍵詞：正則化方法下生存模型的個(gè)人信用風(fēng)險(xiǎn)分析　出處：《上海師范大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 信用風(fēng)險(xiǎn) 生存分析 變量正則化 Logistic回歸 決策樹

【摘要】：信用風(fēng)險(xiǎn)是銀行業(yè)的一個(gè)關(guān)鍵領(lǐng)域,是機(jī)構(gòu)、消費(fèi)者和監(jiān)管機(jī)構(gòu)等各種利益相關(guān)者共同關(guān)注的問題。信用風(fēng)險(xiǎn)的研究是金融領(lǐng)域的熱點(diǎn)研究主題,近些年也引起了統(tǒng)計(jì)研究者的關(guān)注。Wikipedia(2017)將信用風(fēng)險(xiǎn)定義為:由于債務(wù)人不支付貸款而造成的損失風(fēng)險(xiǎn)或其他信貸額度。信用風(fēng)險(xiǎn)的核心是違約事件,當(dāng)債務(wù)人不能根據(jù)債務(wù)合同償付相關(guān)債務(wù)、履行法定義務(wù),就發(fā)生了違約事件。在銀行客戶信用風(fēng)險(xiǎn)研究中,僅通過客戶是否違約來評價(jià)其信用好壞是不夠準(zhǔn)確的。因?yàn)榇蟛糠挚蛻粼谘芯科趦?nèi)不會(huì)發(fā)生違約行為,我們無法觀測到大部分個(gè)體的生存時(shí)間,這就產(chǎn)生了生存分析中常見的右刪失數(shù)據(jù)。在最近這些年,一些研究將生存分析的方法運(yùn)用到信用風(fēng)險(xiǎn)分析模型中。生存分析是一種動(dòng)態(tài)分析方法,它不僅能預(yù)測事件發(fā)生的概率,也能預(yù)測事件發(fā)生的時(shí)間。它擅長處理刪失數(shù)據(jù)和截尾數(shù)據(jù),利用估計(jì)的生存概率可以更加直觀地反應(yīng)風(fēng)險(xiǎn)與特征因素之間的關(guān)系。同時(shí)在模型中引入時(shí)間變量,能更好的體現(xiàn)對象的生存狀態(tài)。本文基于三年(36期)研究期內(nèi)60508個(gè)樣本銀行客戶420個(gè)高維特征變量的小額貸款脫敏數(shù)據(jù),在傳統(tǒng)的變量選擇方法受到挑戰(zhàn)的情況下,首先對當(dāng)今熱點(diǎn)的正則化方法進(jìn)行查閱比較和算法嘗試。接著,我們創(chuàng)新性的將違約的跨度時(shí)間考慮到信用分析模型中,引入客戶首次違約的期數(shù),將數(shù)據(jù)處理為生存數(shù)據(jù)的固定格式,并分別建立基于LASSO-MCP正則化方法的Cox乘法危險(xiǎn)率模型和基于LASSO-SCAD正則化方法的加法危險(xiǎn)率模型。同時(shí),我們將重要變量的系數(shù)估計(jì)值與對應(yīng)特征變量取值的乘積作為信用得分,建立分類規(guī)則,綜合評價(jià)每一個(gè)客戶的信用風(fēng)險(xiǎn)。通過與銀行業(yè)務(wù)經(jīng)驗(yàn)結(jié)果的反饋對比,給出基于生存模型的部分重要特征變量的經(jīng)濟(jì)意義。最后,我們從重要特征變量的結(jié)果和模型的預(yù)測效果兩個(gè)方面對生存分析的兩個(gè)模型進(jìn)行比較。發(fā)現(xiàn)基于LASSO-MCP正則化方法的比例風(fēng)險(xiǎn)模型用更少的特征變量卻得到了相對更好的分類效果。本文在最后從多個(gè)角度對基于不同方法的信用風(fēng)險(xiǎn)分析模型進(jìn)行效果驗(yàn)證和比較。首先,基于實(shí)證數(shù)據(jù)分別實(shí)現(xiàn)傳統(tǒng)二分類Logistic回歸模型和現(xiàn)代決策樹模型。接著,將前述章節(jié)中生存分析的乘法模型和加法模型與二者比較。基于理論分析和模型結(jié)果,從解釋模型準(zhǔn)確性的ROC曲線和代表模型區(qū)分能力的KS統(tǒng)計(jì)量兩個(gè)方面比較四個(gè)模型,發(fā)現(xiàn)生存分析Cox模型均優(yōu)于其他三種模型。這就從多方面驗(yàn)證了本文引入生存時(shí)間并基于正則化方法建立的生存分析模型的良好實(shí)證效果。從模型整體的準(zhǔn)確性和區(qū)分力兩個(gè)方面,綜合得出:對于三年期小額貸款數(shù)據(jù),基與LASSO-MCP正則化方法的生存分析Cox比例風(fēng)險(xiǎn)模型有最高的準(zhǔn)確性和最大的模型區(qū)分力。
[Abstract]:Credit risk is a key area of banking, is a common concern of various stakeholders, such as institutions, consumers and regulators. The research of credit risk is a hot research topic in the field of finance. In recent years it has also attracted the attention of statisticians. Wikipedia2017). The credit risk is defined as the loss risk or other credit line caused by the debtor's failure to pay the loan. The core of the credit risk is the default event. When the debtor can not pay the related debt according to the debt contract, and fulfill the legal obligations, there is a default event. In the study of the credit risk of bank customers. It is not accurate to judge the credit quality of customers simply by whether they default or not, because most customers do not default during the study period, and we can not observe the survival time of most individuals. In recent years, some studies have applied the method of survival analysis to credit risk analysis model. Survival analysis is a dynamic analysis method. It not only can predict the probability of the event, but also can predict the time of the event. It is good at dealing with censored data and censored data. The estimated survival probability can reflect the relationship between risk and characteristic factors more intuitively. At the same time, time variables are introduced into the model. This paper based on 60508 sample bank customers during the research period 420 high-dimensional characteristic variables of micro-credit desensitization data. When the traditional method of variable selection is challenged, the regularization methods of today's hot spots are first compared and the algorithms are tried. We creatively take the span of default into account of the credit analysis model, introduce the number of customer first default period, and process the data into a fixed format of survival data. Cox multiplicative hazard rate model based on LASSO-MCP regularization method and additive hazard rate model based on LASSO-SCAD regularization method are established respectively. We take the product of coefficient estimate of important variable and the value of corresponding characteristic variable as credit score and establish classification rules. Comprehensive evaluation of the credit risk of each customer. By comparing with the results of bank experience, the economic significance of some important characteristic variables based on survival model is given. Finally. We compare the two models of survival analysis in terms of the results of important feature variables and the prediction effect of the model. It is found that the proportional risk model based on LASSO-MCP regularization method uses fewer features. In the end, this paper validates and compares the credit risk analysis model based on different methods from several angles. Based on the empirical data, the traditional two-classification Logistic regression model and the modern decision tree model are implemented respectively. The multiplication model and addition model of survival analysis in the previous chapters are compared with the two models, based on theoretical analysis and model results. The four models are compared from two aspects: the ROC curve which explains the accuracy of the model and the KS statistics which represent the distinguishing ability of the model. It is found that the survival analysis Cox model is superior to the other three models, which verifies the good empirical effect of the survival analysis model introduced in this paper based on the regularization method. There are two aspects: accuracy and differentiability. It is concluded that for three-year microfinance data, the Cox proportional risk model has the highest accuracy and maximum distinguishing power between the base and the LASSO-MCP regularization method.
【學(xué)位授予單位】：上海師范大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：F832.4

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 葉永剛;吳良順;;基于BP神經(jīng)網(wǎng)絡(luò)模型的創(chuàng)業(yè)板上市公司信用級別評估和信用風(fēng)險(xiǎn)度量[J];經(jīng)濟(jì)與社會(huì)發(fā)展;2016年03期

2 李從剛;童中文;曹筱玨;;基于BP神經(jīng)網(wǎng)絡(luò)的P2P網(wǎng)貸市場信用風(fēng)險(xiǎn)評估[J];管理現(xiàn)代化;2015年04期

3 內(nèi)蒙古銀行課題組;楊海平;陳晶晶;;基于logistic回歸的小微企業(yè)信用風(fēng)險(xiǎn)預(yù)警[J];內(nèi)蒙古金融研究;2014年08期

4 龐素琳;鞏吉璋;;C5.0分類算法及在銀行個(gè)人信用評級中的應(yīng)用[J];系統(tǒng)工程理論與實(shí)踐;2009年12期

5 易傳和;彭江;;基于FAHP的個(gè)人信用評分模型[J];統(tǒng)計(jì)與決策;2009年15期

6 張成虎;李育林;吳鳴;;基于判別分析的個(gè)人信用評分模型研究與實(shí)證分析[J];大連理工大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2009年01期

7 李曉卉;;決策樹技術(shù)在客戶信用分析中的應(yīng)用[J];武漢科技大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2008年02期

8 余文建;沈益昌;杜洋;;基于Logistic模型的個(gè)人信用評分體系研究[J];海南金融;2007年03期

9 陳忠陽;違約損失率(LGD)研究[J];國際金融研究;2004年05期

相關(guān)博士學(xué)位論文 前2條

1 付光輝;高維的強(qiáng)相關(guān)數(shù)據(jù)的模型選擇[D];中南大學(xué);2011年

2 錢俊;生存分析中刪失數(shù)據(jù)比例對Cox回歸模型影響的研究[D];南方醫(yī)科大學(xué);2009年

相關(guān)碩士學(xué)位論文 前6條

1 張丹婷;基于生存分析的信用風(fēng)險(xiǎn)量化研究[D];浙江大學(xué);2015年

2 陳麗;上市公司信用風(fēng)險(xiǎn)評價(jià)的Fisher判別分析模型[D];重慶大學(xué);2013年

3 張s，

本文編號：1425758