Дисертації з теми "Prediction of loan default"
Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Prediction of loan default.
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Prediction of loan default".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Granström, Daria, and Johan Abrahamsson. "Loan Default Prediction using Supervised Machine Learning Algorithms." Thesis, KTH, Matematisk statistik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-252312.
Повний текст джерелаАнотація:
It is essential for a bank to estimate the credit risk it carries and the magnitude of exposure it has in case of non-performing customers. Estimation of this kind of risk has been done by statistical methods through decades and with respect to recent development in the field of machine learning, there has been an interest in investigating if machine learning techniques can perform better quantification of the risk. The aim of this thesis is to examine which method from a chosen set of machine learning techniques exhibits the best performance in default prediction with regards to chosen model evaluation parameters. The investigated techniques were Logistic Regression, Random Forest, Decision Tree, AdaBoost, XGBoost, Artificial Neural Network and Support Vector Machine. An oversampling technique called SMOTE was implemented in order to treat the imbalance between classes for the response variable. The results showed that XGBoost without implementation of SMOTE obtained the best result with respect to the chosen model evaluation metric.Det är nödvändigt för en bank att ha en bra uppskattning på hur stor risk den bär med avseende på kunders fallissemang. Olika statistiska metoder har använts för att estimera denna risk, men med den nuvarande utvecklingen inom maskininlärningsområdet har det väckt ett intesse att utforska om maskininlärningsmetoder kan förbättra kvaliteten på riskuppskattningen. Syftet med denna avhandling är att undersöka vilken metod av de implementerade maskininlärningsmetoderna presterar bäst för modellering av fallissemangprediktion med avseende på valda modelvaldieringsparametrar. De implementerade metoderna var Logistisk Regression, Random Forest, Decision Tree, AdaBoost, XGBoost, Artificiella neurala nätverk och Stödvektormaskin. En översamplingsteknik, SMOTE, användes för att behandla obalansen i klassfördelningen för svarsvariabeln. Resultatet blev följande: XGBoost utan implementering av SMOTE visade bäst resultat med avseende på den valda metriken.
2
Stone, Devon. "An exploration of alternative features in micro-finance loan default prediction models." Master's thesis, Faculty of Science, 2020. http://hdl.handle.net/11427/32377.
Повний текст джерелаАнотація:
Despite recent developments financial inclusion remains a large issue for the World's unbanked population. Financial institutions - both larger corporations and micro-finance companies - have begun to provide solutions for financial inclusion. The solutions are delivered using a combination of machine learning and alternative data. This minor dissertation focuses on investigating whether alternative features generated from Short Messaging Service (SMS) data and Android application data contained on borrowers' devices can be used to improve the performance of loan default prediction models. The improvement gained by using alternative features is measured by comparing loan default prediction models trained using only traditional credit scoring data to models developed using a combination of traditional and alternative features. Furthermore, the paper investigates which of 4 machine learning techniques is best suited for loan default prediction. The 4 techniques investigated are logistic regression, random forests, extreme gradient boosting, and neural networks. Finally the paper identifies whether or not accurate loan default prediction models can be trained using only the alternative features developed throughout this minor dissertation. The results of the research show that alternative features improve the performance of loan default prediction across 5 performance indicators, namely overall prediction accuracy, repaid prediction accuracy, default prediction accuracy, F1 score, and AUC. Furthermore, extreme gradient boosting is identified as the most appropriate technique for loan default prediction. Finally, the research identifies that models trained using the alternative features developed throughout this project can accurately predict loan that have been repaid, the models do not accurately predict loans that have not been repaid.
3
Shan, Chenyu, and 陜晨煜. "Credit default swaps (CDS) and loan financing." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2013. http://hub.hku.hk/bib/B5089965X.
Повний текст джерелаАнотація:
As evidenced by its market size, credit default swaps (CDSs) has been the cornerstone product of the credit derivatives market. The central question that I attempt to answer in this thesis is: why and how does the introduction of CDS market affect bank loan financing? Theoretical works predict some potential effects from CDS market, but empirical evidence is still rare.
This dissertation empirically examines the effects of CDS trading on bank loan financing. In chapter one, I find that banks increase average loan amount and charge higher loan spread after the onset of CDS trading on the borrower’s debt. Also, credit quality of the borrower deteriorates for those with active CDS trading. These findings suggest that banks tend to take on more credit risk by issuing larger loans and by lending to riskier firms that could not obtain bank loan in the absence of CDS. The risk-taking by banks ultimately transmitted to higher bank-level risk profile.
The second chapter is the first empirical study of CDS’ role in determining loan syndicate structure. I find larger lead bank share when CDS is in place. Moreover, participation of credit derivatives trading by lead banks is much larger than by the participants, suggesting that lead banks have better chance to use CDS to their own advantage. Further analysis shows that lead banks retain an even larger share when it is more experienced dealing with the borrower and when information asymmetry between the lender and the borrower is less severe. Different from conventional wisdom about moral hazard in syndicated lending, our findings suggest that the lead bank likely takes on more credit risk voluntarily due to its increased financing capacity.
The third chapter focuses on the effects of CDS on debt contracting. Given that current evidence does not show CDS reduces average cost of debt, we conjecture that the diversification benefit is reflected by relaxation of restrictions imposed on borrowers. Consistent with our hypothesis, we find the marginal effect from CDS trading on covenant strictness measure is 16.8% on average. One standard deviation increase in the number of outstanding CDS contracts loosens net worth covenants by approximately 8.9%. Using various endogeneity controls, we are able to show the loosening of covenants is due to the reduced level of debtholder-shareholder conflict. Furthermore, the loosening effect is stronger when the expected renegotiation cost is larger, consistent with the view that CDS mitigates contracting friction and improves contracting efficiency.
Overall, this dissertation attempts to provide first empirical evidence on how CDS affects bank loan financing. We focus the analysis on loan issuance, syndicate structure and contracting. The findings suggest that banks lend to riskier borrowers in the presence of CDS. On a positive note, banks tend to impose less restrictive covenants on its borrower, which may mitigate frictions in lending market in terms of ex ante bargaining and ex post renegotiation cost.published_or_final_version
Economics and Finance
Doctoral
Doctor of Philosophy
4
Tian, Shaonan. "Essays on Corporate Default Prediction." University of Cincinnati / OhioLINK, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1352403546.
Повний текст джерела
5
Olsen, Hunter. "An Analysis of Post Great Recession Student Loan Default." Scholarship @ Claremont, 2018. http://scholarship.claremont.edu/cmc_theses/1960.
Повний текст джерелаАнотація:
With more than $1.48 trillion in outstanding student loans and nearly five million Americans in default in 2017, student loans may pose one of the greatest threats to financial stability of individuals in the coming years. Failing to pay loans on time may result in wage garnishment and the suspension of Social Security payments. The second largest form of household debt, student loans are almost never dischargeable in bankruptcy and yet are critical for millions to make investments in human capital. This thesis utilizes the October 2017 addition of administrative data in the Beginning Postsecondary Students (BPS) to analyze factors influencing likelihood of student loan default in the United States up to 20 years post-enrollment. It applies logistic regression analysis to BPS 1996 and BPS 2004 and is able to trace the evolution of contributing factors over time.
6
Leow, Mindy. "Credit risk models for mortgage loan loss given default." Thesis, University of Southampton, 2010. https://eprints.soton.ac.uk/170515/.
Повний текст джерелаАнотація:
Arguably, the credit risk models reported in the literature for the retail lending sector have so far been less developed than those for the corporate sector, mainly due to the lack of publicly available data. Having been given access to a dataset on defaulted mortgages kindly provided by a major UK bank, this work first investigates the Loss Given Default (LGD) of mortgage loans with the development of two separate component models, the Probability of Repossession (given default) Model and the Haircut (given repossession) Model. They are then combined into an expected loss percentage. Performance-wise, this two-stage LGD model is shown to do better than a single-stage LGD model (which directly models LGD from loan and collateral characteristics), as it achieves a better Rsquare value, and it more accurately matches the distribution of observed LGD. We next investigate the possibility of including macroeconomic variables into either or both component models to improve LGD prediction. Indicators relating to net lending, gross domestic product, national default rates and interest rates are considered and the interest rate is found to be most beneficial to both component models. Finally, we develop a competing risk survival analysis model to predict the time taken for a defaulted mortgage loan to reach some outcome (i.e. repossession or non-repossession). This allows for a more accurate prediction of (discounted) loss as these periods could vary from months to years depending on the health of the economy. Besides loan- or collateral-related characteristics, we incorporate a time-dependent macroeconomic variable based on the house price index (HPI) to investigate its impact on repossession risk. We find that observations of different loan-to-value ratios at default and different security type are affected differently by the economy. This model is then used for stress test purposes by applying a Monte Carlo simulation, and by varying the HPI forecast, to get different loss distributions for different economic outlooks.
7
Ngufor, Patrick. "Quantitative study of perceptions of business owners and loan officers on loan delinquency and default." Thesis, University of Phoenix, 2014. http://pqdtopen.proquest.com/#viewpdf?dispub=3578584.
Повний текст джерелаАнотація:
This research seeks to document if differences in perceptions of small business creditworthiness and lending practices of credit union and commercial lenders exist. This study applied a quantitative method to answer five questions: (1)How do small business owners perceive commercial lenders? (2)How do small business owners perceive credit union lenders? (3)How do commercial banks perceive small businesses? (4)How do credit unions perceive small businesses? (5)What are the differences in the perceptions of small businesses, commercial banks, and credit unions? The study used a quantitative survey instrument to gather data and the data was compared and contrasted among groups (Fitzgerald & Rumrill, 2004). The chi-squared test of differences in probabilities and the goodness of fit test were applied (Figure 2A) to determine if there were differences in probabilities between answers.
The results of this study are significant to small business and banking leaders by helping to define how lenders’ and small businesses’ perceptions affect the differences in loan delinquency rates between commercial lenders and credit unions lenders and by offering new insight into how loan delinquency rates can be reduced. The results also pointed to inherent perceptions of small business owners and lenders that might contribute to the root causes of loan defaults and delinquencies. The results provided information upon which small business owners and financial institution loan officers might act in order to understand how to better manage loans and to reduce the rate of loan delinquency.
8
Aguilera, Nelson A. "Credit rationing and loan default in formal rural credit markets." Connect to resource, 1990. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1232115721.
Повний текст джерела
9
Mphaka, Patrick. "Strategies for Reducing Microfinance Loan Default in Low-Income Markets." ScholarWorks, 2017. https://scholarworks.waldenu.edu/dissertations/4391.
Повний текст джерелаАнотація:
Poor loan repayment causes the decline and failure of some microfinance institutions. The purpose of this qualitative multiple case study was to explore strategies that microfinance (MFI) leaders use to reduce loan default in the base of the pyramid market. The study population included 6 MFI leaders, 12 borrower community-based groups, and 4 staff members of the Adventist Development and Relief Agency (ADRA Rwanda) who reduced MFI loan default in Rwanda. Data were collected through semistructured interviews with 3 MFI leaders, 3 ADRA Rwanda staff members, and 3 members of borrower groups. Data were also collected through focus groups with 3 borrower community-based groups comprising 6 to 8 members. Additional data were collected through the analysis of MFI and ADRA Rwanda organizational documents. The Varian group lending model was the conceptual framework for the study. Data analysis involved methodological triangulation and the Gadamerian hermeneutics framework of interpretation. Four major themes emerged: intrapreneurship and environmental business opportunities, favorable loan repayment conditions, strategies for choosing borrower groups, and loan use monitoring. A sustainable microfinance institution can produce social change by providing microfinance loans that clients can use to start and grow microenterprises that can become the source of income for improving the lives of clients and their family members. Findings may also be used to create economic growth through the participation of more people in economic activities in the base of the pyramid market.
10
Wang, Yangzhengxuan. "Corporate default prediction : models, drivers and measurements." Thesis, University of Exeter, 2011. http://hdl.handle.net/10036/3457.
Повний текст джерелаАнотація:
This thesis identifies the optimal set of corporate default drivers and examines the prediction performance of corporate default measurement tools, using a sample of companies in the United States from 1970 to 2009. In the discussion of optimal default drivers, feature selection techniques including the t-test and stepwise methods are used to filter relevant default information collected from previous empirical studies. The optimal default driver information set consists of quantitative parameters from accounting ratios, market indices, macroeconomic indicators, default history, and firm age. While both accounting ratios and market information dominate the explanatory ability, followed by default history, macroeconomic indicators contribute additional explanation for default risk. Moreover, industry effects show significance across alternative models, with the retail industry presenting as the sector with highest risk. The results are robust in both traditional and advanced random models. In investigating the optimal prediction method, two newly developed random models, mixed logit and frailty model, are tested for their theoretical superiority in capturing default clusters and unobservable information for default risk. The prediction ability of both models has been improved upon using the extended optimal set of default drivers. While the mixed logit model provides better prediction accuracy and shows stability in robustness checks, the frailty model benefits from computational efficiency and explains default clusters more thoroughly. This thesis further compares the prediction performance of large dimensional models across five categories based on the default probabilities transferred from alternative results in different models. Besides the traditional assessment criteria - covering the receiver operating characteristic curve, accuracy ratios, and classification error rates – this thesis thoroughly evaluates forecasting performance using innovative proxies including model stability under financial crisis, profitability and misclassification costs for creditors using alternative risk measurements. The practical superiority of the two advanced random models has been verified further in the comparative study.
11
Emenike, Obioma. "Business loan default in Nigerian commercial banks : from causes to remedies." Thesis, Stellenbosch : Stellenbosch University, 2011. http://hdl.handle.net/10019.1/97167.
Повний текст джерелаАнотація:
Thesis (MDF)--Stellenbosch University, 2011.ENGLISH ABSTRACT: A sound and favourable financial climate is necessary for any forward-looking economy to thrive. This, amongst others, includes the extent to which the commercial banks are able to discharge their intermediating role in the demand and supply of credit necessary to sustain commercial businesses. Indeed, in the last decade, the Nigerian banking industry has witnessed swings with the attendant effects on the business community. One of the downsides has been the incidence of loan default which led to many banks recording astronomical levels of bad loans in their 2008 financial reports. The drastic measures taken by the Central Bank of Nigeria of relieving eight CEOs of their jobs in September 2009 further highlights the import of this subject matter. This paper gives an overview of the concept of loan default in Nigerian commercial banks ranging from the causes to the remedies currently in place to checkmate it. A field survey on loan officers, credit analysts and credit risk managers in some select banks was carried out. The findings reveal that the banks have a rather cautious approach to lending with certain classes of loans classified. Causal factors leading to loan delinquencies categorised into environmental, bank specific and borrower specific factors were analysed to have contributed equally to causing loan default in Nigeria. Lastly, the regression results indicated that there was a significant relationship between measures adopted by the banks in the face of increa
12
Kelley, Samuel Hanson. "Factors Affecting Student Loan Default in Proprietary Non-Degree Granting Colleges." ScholarWorks, 2017. https://scholarworks.waldenu.edu/dissertations/3898.
Повний текст джерелаАнотація:
The significant problem addressed in this research was the increasing default rate among federal student loan borrowers who attended non-degree-granting proprietary colleges in Florida (i.e., career and technical colleges). The purpose of this study was to identify, better understand, and predict which borrower characteristics increased the likelihood of student loan default at proprietary non-degree-granting colleges. The research was based on the structural-functional and planned behavior theories and utilized a quantitative, non-experimental, cross-sectional design to explore the relationship between academic success, age, college graduation status, ethnicity, gender, high school class ranking, and federal student loan default. Self-reported data were obtained from students who attended private, for-profit, less than 2-year colleges in Florida. To determine which student borrower characteristics predicted an increase in the likelihood that borrowers would default on their student loan payments, one hypothesis was proposed to evaluate six borrower characteristics. Logistic regression analysis was used to explore the statistical relationships and found that academic success, age, and gender were statistically significant in predicting student loan default among students who attended private, for-profit, less than 2-year colleges in Florida. This study may facilitate positive social change by aiding educational institutions in identifying at-risk borrower characteristics and by providing various default prevention strategies that could be incorporated into specific counseling messages to reduce future student loan defaults and lower institutional cohort default ratings.
13
Black, Kevin. "Determining capital adequacy for a community bank's agricultural loan portfolio." Thesis, Kansas State University, 2015. http://hdl.handle.net/2097/35221.
Повний текст джерелаАнотація:
Master of AgribusinessDepartment of Agricultural Economics
Brian C. Briggeman
As the recent financial crisis brought to light, the ability of commercial banks to quantify and better manage risk in their loan portfolios is paramount to their continued success and viability. Assessing, managing, and retaining capital is now a larger issue than ever given this event as well as the advent of the Basel III Accord. Pinnacle Bancorp is a community banking organization headquartered in Omaha, Nebraska with roughly $8.6 billion in assets. The company is also one of the largest agricultural lenders in the country and the largest agricultural lender among traditional community banks. Given the ominous outlook heading into 2016 for agricultural producers from lower projected net incomes and increased borrowing costs following Federal Reserve action on the Fed Funds Rate, many banks worry about the increased likelihood of default for agricultural producers. The objective of this thesis is to determine the adequacy of Pinnacle Bank’s equity capital relative to the agricultural loan portfolio. This process begins by employing binary logit regression in an effort to determine the probability of default for the bank’s agricultural loan portfolio. With default likelihood quantified, efforts are then made to determine the bank’s credit value-at-risk at various solvency levels. These figures are then compared to current capital levels in order to determine the adequacy of bank capital as measured by five key regulatory ratios ultimately imposed by Basel III. Finally, recommendations are made to management as to the adequacy of bank capital relative to the agricultural loan portfolio and any future efforts that need to be made in order to determine and ensure the adequacy of bank capital for the entire loan portfolio.
14
Lai, Hengwa. "Agency risk in CMBS default resolution : a case study of the Peter Cooper Village - Stuyvesant Town mortgage loan default." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/62055.
Повний текст джерелаАнотація:
Thesis (S.M. in Real Estate Development)--Massachusetts Institute of Technology, Program in Real Estate Development in Conjunction with the Center for Real Estate , 2010.Cataloged from PDF version of thesis.
Includes bibliographical references (p. 54-56).
Between 2010 and 2018, approximately $410 billion of maturing CMBS loans are expected not to able to refinance; that is, they are in high risk of default. The current real estate downturn has not only pushed delinquencies to a historic high but has also inflicted losses to bondholders. When losses are realized through foreclosure, junior bondholders can have the face amount of their investment significantly reduced with no cash payment, while the senior bondholders receive partial repayment of their investment at par. Alternatively, loan modifications, or workouts, yield different outcomes which are more favorable to the junior bondholders. The rising tide of loan defaults and loan workouts will certainly exacerbate the ongoing "tranche war" among the CMBS bondholders. Consequently, it is imperative to understand how the CMBS servicing structure governs default resolution and loan workouts. By analyzing the recent default of Peter Cooper Village-Stuyvesant Town, this study will examine the case of the largest commercial real estate default in the US history as a real life example to illustrate whether the overlapping role of B-Piece buyer and Special Servicer adversely affects workout prudence. Through interviews with industry professionals and a review of the Pooling and Servicing Agreement (PSA), and a review of the transcript of the CMBS Investment Grade Bondholder Forum in June, 2010, the study proposes structural changes that could potentially mitigate agency risk inherent in the current servicing structure..
by Hengwa Lai.
S.M.in Real Estate Development
15
Chan, Yuen-yee Emily. "A Study of mortgage transaction goverance in Hong Kong with particular reference to mortgage default." Click to view the E-thesis via HKU Scholars Hub, 2004. http://lookup.lib.hku.hk/lookup/bib/B37905284.
Повний текст джерела
16
Betz, Jennifer [Verfasser], and Daniel [Akademischer Betreuer] Rösch. "Resolution of defaulted loan contracts - An empirical analysis of default resolution time and loss given default / Jennifer Betz ; Betreuer: Daniel Rösch." Regensburg : Universitätsbibliothek Regensburg, 2018. http://d-nb.info/1164765604/34.
Повний текст джерела
17
Sustersic, Jennifer Lynn. "Do traded credit default swaps impact lenders' monitoring activities? Evidence from private loan agreements." The Ohio State University, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=osu1339512002.
Повний текст джерела
18
Chou, Hsin-Yi, and 周欣怡. "Mortgage Loan Default Prediction-The Application of Survival Analysis Model." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/67031033284360611934.
Повний текст джерелаАнотація:
碩士真理大學
財經研究所
96
Mortgage loan is one of the main earnings of the Banking, and residential mortgage loan takes the major part of consumer loan business. In recent years, there are competing intensely between the banks. In order to have an occupation in the market, banks forsake their credit principles. Due to the loan quality became worse and worse, and non-performing loan ratio remains high. Such as dual-card storm (credit cards and cash cards) in 2005 and American sub-prime mortgages in 2007, all of they have serious influence in financial market. So as to diminish non-performing loan problem, banks should to construct an effective housing loan credit scoring system. Therefore banks can grasp most correct background information of creditor, and reduce the default risk in the future. This paper based on a local commercial bank’s housing loan databank from November 1, 1995 to November 30, 2005. We apply Survival Analysis-Cox regression Model, Logistic Regression Model, CHAID decision tree model and CART decision tree model to construct the Mortgage Loan Scoring model. Results show that age, occupation...variables have an significance to normal and default loans. The empirical results also show that Cox model can outperform the Logit, CHAID and CART model in percentage of correct rate in out-sample. This represent the Cox model has a high performance than Logit, CHAID and CART model. In the last, to compare the performance of out-sample, we consider the ROC ratio. The results also reveal that Cox model can still beat the entire classified model.
19
CHIU, CHING-HUAN, and 邱敬桓. "An Application of Feature Selection to P2P Loan Default Prediction." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/y7ym8q.
Повний текст джерелаАнотація:
碩士輔仁大學
金融與國際企業學系金融碩士班
107
Along with the trend of the development of the financial technology，online peer to peer is growing rapidly，the necessity of the risk control on P2P is more and more important issue，hence this study uses the machine learning to analyze the data of the Lending Club，which is the biggest P2P platform in America and to predict P2P loan default. Feature selection is a pretty crucial procedure before analyzing the data，because it not only can reduce the complexity of the data construction，but avoid the occurrence of overfitting. This study chooses the random forest、extreme gradient boosting、logistic regression and support vector machine as the feature selection model. Finally, this study gets the intersection according to the result of the four models，so eight features have been chosen，such as the ratio of total current balance to high credit、debt to income、funded amount、grade、installment、home ownership、purpose and total high credit/credit limit. Then, we choose the extreme gradient boosting as our classification model，the result shows that the power of prediction has been improved，and this study selects four indicators to stand for the greater prediction，such as “Accuracy”、“Precision”、“F1 score” and “G means”. For the P2P platform，the result of this study not only can reduce the cost when collecting the lending factors，but lower the situation of default.
20
Chen, Jie-Awei, and 陳皆回. "The Performance Comparison of Various Default Prediction Models for Mortgage Loan." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/09010428489538842045.
Повний текст джерелаАнотація:
碩士朝陽科技大學
財務金融系碩士班
99
The mortgage loan is the highest proportion of various consumptive loan services for Taiwan financial institutions, includes domestic banks, the local branches of foreign banks and medium business banks. The revenues of the Taiwan financial institutions is significantly related to the proportion of overdue mortgages. With gradual opening of financial policy, financial system heads for liberalization and globalization. Local banks, private or public, intend to increase market share so that the crediting principles are somewhat loose and quality lowers as well. As a result, the bad debt rate is on the increase year by year. This study attempts to Classification tree model, Bayesian classification and the rough set model as a supplementary tool to explore the predictive value, and to identify more representative decision-making model. The purpose of this study is to build a fair and objective default prediction models for mortgage loan. The findings of this study can provide useful information to banks for making decision about credit policy of personal mortgage in the future. Besides, it can increase the correctness from sieve of default customer and decrease the loss of banks. The empirical results show that Classification tree model has the highest overall accuracy rate reached 85.3%.
21
Yeh, Tse-Ying, and 葉澤瑩. "An Application of Survival Analysis to P2P Lending Loan Default Prediction." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/3ajc5c.
Повний текст джерелаАнотація:
碩士輔仁大學
金融與國際企業學系金融碩士班
107
P2P Lending is at higher default risk than traditional financial institutions most of the time. In this research, we use Cox proportional hazards model in survival analysis to study P2P lending loan default and analyze effects of personal financial conditions and environmental fluctuation by 10 individual and 5 macroeconomic factors. We hope to predict the timing of default on P2P lending. In this research, we use loan data on Lending Club in 2012~2015. The observation period is from Jan, 2012 to Oct, 2018. We define "default" as payments which are 120 days overdue, and "survival period" as months from lending to default. The results show that the lower the default cutoff point is, the higher the accuracy of loan status and the higher the missed loan ratio (numbers of loans predicted normal but actually default / numbers of loans predicted normal) is. We find that a shorter predicted period outperform a longer period. Such results are not sensitive to the choices of cutoff points for the predicted period of one and a half to two years. For survival period prediction, the higher the cutoff point is, the better the early warning effect is. Also, the missed loan ratio and survival period predictability will be better when including macroeconomic factors. Besides, the survival curves of the following factors have high sensitivity: interest rate, annual income, debt to income, inquiries in past 6 months, grade, home ownership, and purpose.
22
Ma, Jo-Ya, and 馬若雅. "The Influence of Open Data on the P2P Loan Default Prediction." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/36r7hy.
Повний текст джерелаАнотація:
碩士輔仁大學
金融與國際企業學系金融碩士班
107
P2P lending industry is gradually emerging in Taiwan. Due to the lack of traditional analytical data, it is necessary to collect new data to assist in the analysis. The open data is relatively easy to collect, and the collection cost is low. The study will test the influence of open data on the p2p loan default prediction. We used the Prosper platform as the research object. During the study period, from 2010 to 2014, the regression prediction model was established using logistic regression and random forest. In addition to the control variables, the explanatory variables are added to the control variable model by public data, and the other part is used as the basis for segmentation data, and the predicted default rate after grouping is compared with the default rate of the parent. From the empirical results, it can be known that the data will be higher than the maternal default rate, with a rate of about 60%, and the predicted default rate can be increased by up to 10%. Therefore, the study believes that the open data variable can improve the existing default forecast.
23
Yi, Jessica Moonhee. "Identification of relevant predictors of loan default using the Elastic Net model." Thesis, 2017. http://hdl.handle.net/2440/114273.
Повний текст джерелаАнотація:
The timely prediction of loan default plays an important role in lending decisions and monitoring loans. However, there has been little development of models for the selection of relevant variables for the prediction of loan default. This study identifies financial and economic indicators for the forward-looking prediction of loan default by the application of a penalised regression approach, namely the Elastic Net model.
The study employs a sample of US firms with 162 loan default events in total between 1998 and 2013. The sample is sub-divided to form a Test sample and two holdout samples: one drawn from the same period as the Test sample; and one drawn from a subsequent period. The sample of non-defaulting firms is constructed using prior probabilities based on the bond default rate for each year.
The 278 potential variables, including the ten economic indicators and 268 financial ratios or summary indicators, are regularised with the application of the Elastic Net model. This process results in the extraction of the ten predictor variables, thus identified as relevant to distinguishing between defaulting and non-defaulting firms. Only one economic indicator, the interest rate, is identified as relevant to the prediction of loan default.
The prediction-usefulness of identified predictor variables are tested using the two most widely used conventional prediction models, multiple discriminant analysis (MDA) and logistic regression (Logit). The resulting MDA and Logit models are compared with Altman’s Z-score model and Ohlson’s O-score model, respectively. Both the Elastic Net prediction models provide more logical explanations of the distinctive characteristics of loan defaulting firms than the Altman’s Z-score and Ohlson’s O-score models. The Elastic Net prediction models outperform the Altman’s Z-score and Ohlson’s O-score models in the accuracy of the Type I, the Type II and the overall classification. When applied to a holdout samples within and outside the same periods, the prediction accuracy of the Elastic Net models is maintained for both defaulting and non-defaulting firms.
This thesis contributes to the loan default literature by introducing the Elastic Net model for variable selection which enhances the predictive ability of the loan default prediction model. The findings of this thesis are potentially useful to financial institutions. Identification of financial and economic predictor variables of loan default can also facilitate assessment of the credit risk of loan applicants. The findings of this thesis may also facilitate better loan default prediction for purposes of monitoring loans. Lastly, the identification of relevant predictor variables may be useful for the classification of loans in the application of the expected loss model in the preparation of financial statements.Thesis (Ph.D.) -- University of Adelaide, Business School, 2018
24
Chuang, Ming-Ter Morris, and 莊明德. "A Study of e-loan Default Prediction Modeling with Neural Network Applications." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/15263358316410250727.
Повний текст джерелаАнотація:
碩士元智大學
管理研究所
97
The purpose of this study is going to set up the financial report early alarm model with Neural Network methodology. The forecast model is used to detect the financial crisis in few years latter happening in a company. Most Enterprises encounter the financial crisis can often be offered from the financial ratio from the financial report. This research will combine each financial ratio from the Taiwan Economic Journal database, through the analytic approach that combines the Self-Organizing Feature Map (SOFM) to find out the similar group, and using K-means looking for fitness clustering. In the process, we then choose the six majors industries in Taiwan, and pick up the 89 companies, which including the listed companies or was in the listed companies to do analysis. Finally, we asked an expert to help us recognizing the company with our clustering levels, and hope the model that is purposed by this research is able to an analysis tool in the practice.
25
YEH, HSIN-YI, and 葉昕宜. "Data Science for Loan Default Probability Prediction in Online Peer-to-Peer Lending." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/ur7f37.
Повний текст джерелаАнотація:
碩士輔仁大學
資訊管理學系碩士班
107
In recent years, P2P lending had become a global trend because of the development of financial technology. P2P lending is a way of crossfunding from the lenders through the Internet, and then loaning the collected funds to the borrower. As an example of Lending Club, the world's largest online P2P lending market. Since 2011, the number of loans and the amount of loans were increase year over year. In 2017, there were 750,000 loans, and the loan amount was reach up to 8.9 billion. In traditional financial institutions, there will be strict review criteria and access to important credit score of the borrower. The P2P platform was designed to eliminate the cumbersome lending process of financial institutions from the huge amount of data collected in the past, then analyzed historical through data exploration. This study will build a credit scoring model through machine learning to eliminate guesswork in financial decisions. This study proposes a data science framework to solve the probability of default in P2P lending based on machine learning. This process included data preprocessing, imbalanced data processing, feature selection, learning algorithm, optimization model hyperparameter, evaluation method and feature importance ranking. In the case of Lending Club, using different imbalance methods to sample features like personal characteristics, credit data, and platform, then use the LASSO algorithm to select the important features. This study created multiple models like logistic regression, neural network, random forest and XGBoost, then find the best hyperparameters for each model using the particle swarm optimization algorithm. Finally, using different metrics to evaluate those models, and find the important features to predict the probability of default of the borrower. This study will demonstrate the feasibility and effectiveness of the P2P credit risk model.
26
Li, Chia-sheng, and 李嘉昇. "The Research of Loan Default Predicting Model for Enterprise." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/08894523089433339897.
Повний текст джерелаАнотація:
碩士淡江大學
財務金融學系碩士在職專班
99
This thesis applies event study to investigate possibility of enterprise default risk. We exam the financial variable and non- financial variable by Z-Score model and Logistic Model from local bank in Taiwan. Corporate lending by banks is also an important source of income for businesses to thrive and growing partner. But the bank''s decision whether to grant business credit line, you must refer to a number of indicators, including the financial dimension and non-financial dimension of the variables. Interest income earned on loans to banks and to avoid loss of corporate defaults,.So how banks in the balance between risk and return, it is a very important issue. This study is based on a domestic commercial bank 97 to 99 years of corporate defaults as the research object households, the use of financial and non-financial dimensions of the variables to do an empirical analysis. Using Altman''s Z-Score value and logistic models Logistic Model to analyze the companies in breach of the annual report is likely to show a default situation, And use normal household financial ratios and default account for comparison to confirm the applicability of the model in the bank''s credit policy.
27
Liou, Tsung-Je, and 劉宗哲. "A Comparative Study of Predictive Models for Mortgage Loan Default." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/95925511748631559903.
Повний текст джерелаАнотація:
碩士國立高雄第一科技大學
金融營運所
90
A Comparative Study of Predictive Models for Mortgage Loan Default Student: Tsung-Je Liou Advisors: Chia-Hsing Huang Department of Financial Operations National Kaohsiung First University of Science and Technology Abstract The debtors’ default on mortgage loan is an important issue for local banks. It does not only deteriorate the capital structure of banks but also damage their profit margin especially during financial crisis or economic depression. Therefore, a default prediction model for banks to improve the mortgage loan decisions and/or to reduce their potential losses is highly needed. This study tries to establish a better prediction model of the debtors’ default for bank’s mortgage loan decision by searching a better combination of predicting variables. Using two statistical techniques (Binary Logistic regression and discriminant analysis), this study analyzes the accuracy rate of prediction and the goodness of fit for two models built in this study and five other models (four from past research and one from the “individual credit evaluation form” of sample bank and then compares their performance with these indicators. In addition, the abilities for all models to minimize misclassification costs are also compared. The result shows that the combination of predicting variables formulated by this study is the dominant model either in the performance indicators or in the ability to minimize the misclassification costs.
28
Chan, Feng-Hua, and 詹鳳華. "Machine Learning Application to Loan Default Comparison and Prediction - A case study in USA and China." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/epykr2.
Повний текст джерелаАнотація:
碩士國立中山大學
國際經營管理碩士學程
106
Online Peer-to-Peer (P2P) lending has been prevailing in the West countries such as USA, UK. The first company to offer peer-to-peer loans in the world was Zopa in UK which provide platform for borrowers can obtain small loan from lenders. In this study, we use machine learning algorithms to predict borrowers’ default risk and discover factors that impact on the rate of loan default with example and data from LendingClub.com. We use logistic regression and random forest to do analysis and identify what is the most influential factor will affect P2P lending. Finally, we compare the difference of P2P lending market between USA and China.
29
CHENG, KAI-TIAN, and 鄭凱天. "Dimension reduction and the performance of peer to peer loan default prediction－Evidence from Lending Club." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/5s74u9.
Повний текст джерела
30
Liu, Chiung-Wen, and 劉瓊文. "A Comparative Study of Predictive Models for Bank Mortgage Loan Default." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/32948830571648911916.
Повний текст джерелаАнотація:
碩士輔仁大學
應用統計學研究所
96
In general, the house mortgage loan is one of the main businesses of banks and the most important income in our banking industry. Therefore, the risk of the banking businesses and the robust management of bank really have the pivotal influence. In 2006, the US sub-prime mortgage crisis has lead to credit crunch. Because the property thriving in Taiwan in the last few years and the domestic banking industry intensely competes, the bank has the loose policy in order to increase banking businesses. The Financial Supervisory Commission, Executive Yuan, announced the uncollected accounts receivable to the cover ratio of Non-Performing Loans was 65.15% at the end of January in 2008. So it is important to build an optional predictive model for the bank. The object of this research is the customers of the house mortgage loan in a bank. The ratio of default is 2.41%. In order to build the model, this study uses the default data and the undefault data to construct the model with the rate 1:2. It repeats 10 times sampling in 1:2 proportion and builds models with Logistic Regression and Cox Regression. Finally, this study builds an optional predictive model for the house mortgage loan. The conclusion is as following: 1. In the method of the study, the optional predictive model is Logistic Regression because the predict index of Logistic Regression is better than Cox Regression. 2. In the analysis of Logistic Regression, the Ensemble model is better than any one of the 10 models. Besides, the optional point cut off the ninth percentile. The accuracy rate is 92.56%, the recall rate is 81.88%, true negative rate is 92.82%, and the precision rate is 21.94%.
31
LIN, FONG SHU, and 馮淑鈴. "The Research of Loan Default Predicting Model for Middle and Small Business." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/70487685411546191557.
Повний текст джерелаАнотація:
碩士中華大學
科技管理學系(所)
96
The middle and small business always play important roles in the track of our country’s development. Because of government policy to facilitate accommodation of middle and small business, the amount of bank loans to middle and small business appears to be increasing. Therefore, for the purpose to enhance bank asset quality, to improve business performance, and to decrease the possibility of middle and small business loan default, it is an important issue to set up a loan default predicting model for banking industry. The object of this research is the middle and small business loan customers of a commercial bank’s branches located in HsinChu and MiaoLio, first we adopt both the financial and non-financial factors to implement an empirical research , and by means of logistic regression to create a predicting model for middle and small business loan risk therewith. According to the result of this research, we find not only the related financial ratios of the default cases vary deeply from year to year, relatively, the related financial ratios of the normal cases are stable, but also some financial ratios show opposite result versus normal status. It reveals the possibility that the default loan customer dresses its financial statement. In terms of non-financial factors, the ratio of default varies from the elements of different industries、period of establishment、collateral、relationship with banks、 age of the person in charge, and whether the person in charge of uses the credit line of cash card or revolving line or not, moreover the difference of default ratio is quite apparent. Finally, this research makes use of the non-financial factors which have discriminating applicability to be the forecasting factors and creates a loan default precaution model for middle and small business thereby, the accurate ratio of the forecast is more than 85%. Hope the result of this research can help financial institutions to set up their middle and small business loan risk evaluation model and adopt some key reference factors therein.
32
Tsai, Shih-Ting, and 蔡仕廷. "The Study on the Optimum Ratio of Oversampling in the Prediction Models of Default Probability of Small Amount Personal Credit Loan." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/37440312343796382859.
Повний текст джерелаАнотація:
碩士輔仁大學
應用統計學研究所
96
The chief revenue of the banks market is the crediting business. The standard of the crediting business may affect the stability of the banks’ operation. Therefore, the banks have to construct the predicting model of default to control the crises. Compared to non-default events, the default events are the rare event. Hence, the process of constructing the default model, the patterns of the default model is always hidden. This study allocates the dataset through the oversampling in order to improve the previous problem. The study uses the small amount personal credit loan (SAPCL) data in a bank to make the research of oversampling. The default rate of the population database is 16.30%. This study uses the sample allocated proportional to the population database to construct the predicting model. On the other hand, oversampling by the different rate of default events and non-default events, which are 1:1, 1:2, 1:3, 1:4, 1:5, are also used. And in order to increase the correct predicting rate, this study also tries two additional rates, which are 3:1 and 2:1. This study evaluates the powers of predictions among 7 different oversampling rates by comparing the results of the predicting models. The conclusions of the study are as follows: In different business strategies, the index which the banks focus on is different. If the bank concerns the true negative rate (TNR), we can use the rate 1:2 to construct the model. If the bank concerns the recall rate (RR), we can use the rate 3:1 to construct the model. If the bank concerns the accuracy rate (AR) and precision rate (RR), we can use the rate 1:5 to construct the model. Because the rate 1:5 is the same as the population database, this study suggests that it does not need to use the oversampling techniques. When considering the cost/profit in the model, the 1;5 oversampling model has the best performance. This means that the oversampling model does not have better performance when considering cost/profit.
33
Lin, Yan-Ping, and 林燕萍. "The Model of Prediction Default Behavior in Consumer Loan—A Comparison of DEA-DA, Neural Networks, Logistic Regression and Discriminant Analysis." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/34226371046996944406.
Повний текст джерелаАнотація:
碩士中華大學
經營管理研究所
95
As the banks focus on the loan to transfer from commercial loan to consumer loan gradually, the banking of consumer loan made boththe banks and the economy growth to bring the effect directly. However increasing the banking of consumer loan simultaneously, it also betrays to the critical problems what the borrower default on loan gradually. Therefore this paper main discussion influence of money attitude toward consumer loan default behavior, and combined the money attitudes and demographic of borrower to establish the pre-warning model for Discriminant analysis, Logistic regression, Neural networks and DEA-DA , developed a set suitable to the pre-warning model for consumer loan defaulted behavior in Taiwan. This study collected the consumer loan data from financial institutions in the Taiwan; we found the money attitudes would affect default behavior throuht the experimental analysis, specially money of attitudes for anxious and maintenance — retention of borrorwer. Moreover, using the basic attributes and money attitudes for immediate information constructed the models, reached above 75% correct rate in four kinds of models, and has the better forecast ability by DEA-DA and neural network model, the sensitivity may reach 80% also the forecast hit rate reaches as high as 94%, this for this article obtained effective pre-warning model. These findings may provide a correct and immediate efficiency pre-warning model for financial industry, also may let the borrower and financial industry understand the importance of money attitudes, raise correct money attitudes improvement expense custom, reduces occurrence the default behavior
34
Yu-ChangHsueh and 薛佑忠. "The Analysis of Credict Risk Elements in Predicting Loan Default for Middle and Small Businesses." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/72246623246711077237.
Повний текст джерелаАнотація:
碩士國立成功大學
財務金融研究所碩士在職專班
103
Based on Altman’s (1968) Z-score model, this study develop a credit risk detecting model for small-to-medium-sized business (SMB) firms by using 163 bank references of 57 SMB firms collected from the loan customers of Tainan branches of a commercial bank during 2011-2013. This study depicts that, in general, when a bank makes a credit assessment of a SMB firm, besides the firm’s financial factors, non-financial factors play an important role in the assessment process as well. While four financial factors appear to be the most frequent used warning indicators, two non-financial factors, “the owner’s educational background” and “the borrower’s and guarantor’s records of repayment of credit card debt or cash card debt”, should be treated as the significant explanatory warning factors. Disturbed by the lack of transparency in the financial reports of SMB firms, this research targets especially on how the non-financial factors enhances bank officers’ understanding of a company’s financial distress to improve the predictive capacity.
35
Wang, Wen-Chun, and 王文崇. "Early Default-warning and the Risk-prediction System of the Bank for SMEs Loans." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/28737385112269070890.
Повний текст джерелаАнотація:
碩士朝陽科技大學
財務金融系碩士班
97
This research mainly applies the factor analytic method and the Logistic regression to construct the bank’s “early default-warning and the risk-prediction system of potential lenders in view of the small and medium-sized enterprises”. Considering the financial structure and management characteristic of “small and medium-sized enterprise”, we design an empirical model which includes 11 explanation variables as well as 4 dummy variables, and carry on the empirical analysis. Again by the Proportional Chance, the Maximum Chance and Press the Q value, we evaluate the accuracy of this system prediction. Furthermore, such as “Bayesian Classification and Classification Tree Model”, to conduct robustness test. In order to help the bank to give the loan reference of the risk profile, the loaning out monitoring and the management before the small and medium-sized enterprise to be the new loan household.
36
Lin, Hsiu-Chuan, and 林秀娟. "Modeling Loss Given Default of Corporate Loan." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/59628343447116972521.
Повний текст джерелаАнотація:
碩士國立清華大學
科技管理研究所
94
The purpose of this article is to construct a loss given default (LGD) estimation model which can discriminate LGD from different characteristics of obligation and collateral, and to offer a reference model for bank that will adopt Advance Internal Rating Based Approach to determine capital requirement of credit risk. This article is based on structured model to construct three different kinds of LGD estimation model which including the loan has no collateral, collateral value is constant and collateral value is stochastic. Specific to the model that concern collateral value is stochastic can take into account the correlation between collateral value and firm’s value, the volatility of collateral value and the volatility of assets value. In empirical analysis we find that the correlation between collateral value and firm’s assets value has significant effect on LGD. Under the same PD, higher correlation between collateral value and firm’s assets value resulting higher LGD. The liquidity of collateral is also an important factor that effect LGD. If the liquidity of collateral is low, it is more difficult to sell collateral to third party and will increase LGD.
37
Wu, Jenq-I., and 吳政毅. "Corporate Governance, Distance from Default, and Loan." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/75275072575725487977.
Повний текст джерелаАнотація:
碩士輔仁大學
會計學系碩士班
95
In this few years, there have been several financial fraud and crisis events occurring, such as the emptying of Procomp Company and Infodisc Company ,the prettified financial statements of Enron and falsely increased yield of World.com. These events lead governments enforced The New Basel Capital Accord in 2006 and corporate governance and risk management becomes hot issues again. Public consider that managers in pursuing personal interests may transfer the interests of the company and shareholders. Therefore, the purpose of the study is going to figure out how to protect the benefits of the shareholders and the relevant parties involved and to ensure the company's operating performance. This study incorporates KMV default risk of development with corporate governance factors. We wish to observe the relation of DD value (Distance of Default), corporate governance, and loan. The results show that these two factors have positive relation with loan amonts. We therefore have a warning suggest for financial demander and supplier that risk management and corporate governance factors will influence performance and finance structure of a company.
38
Liu, Chia-Yi, and 劉家儀. "Time Series Analysis of Loan Default Rate." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/393g2s.
Повний текст джерелаАнотація:
碩士國立交通大學
統計學研究所
107
The use of time series analysis in order to provide decision makers a reference information about future possible changes. The single-dimensional ARIMA (Autoregressive Integrated Moving Average; Box & Jenkins, 1976) model, the GARCH (Generalized Autoregressive Conditional Heteroscedasticity; Bollerslev, 1986) model, and the multidimensional VAR (Vector Autoregressive; Hatemi, 2004) model of time series analysis are mainly fitted the default rate of US loan data set which is expected to achieve good prediction results and comparing the results of different model prediction analysis. Take the default rate of the states of the US loan data from 2008 to 2015 as times series. There are two problems in the data needed to be adjusted. (1) Use differential technique in order to smooth non-stationary sequences. (2) Use GARCH model for solving data which contained heterogeneous variation and different scale characteristics. After the sequence consisting with the time series model assumptions, the prediction models of single-dimensional and multi-dimensional models are constructed. Finally, the prediction abilities of each state's default rate in different modes are compared. The single-dimensional ARIMA model is established under the independence of each state which estimates the variables of the state itself. The multidimensional VAR model considers the relationship between states. At last, observing the prediction efficiency and impact of the single-dimensional and multi-dimensional time series analysis mode. We found a better predictive performance by aggregating information through different states.
39
Hoque, Mohammad Ziaul. "Industrial loan default: the case of Bangladesh." Thesis, 1999. https://vuir.vu.edu.au/15347/.
Повний текст джерелаАнотація:
Industrial development finance institutions (IDFIs) in developing countries have been experiencing serious financial distress since the early 1970s due to persistent loan defaults. Despite the application of a number of remedial measures, industrial loan default problems continued to haunt the IDFIs. The massive loan loss has not only impaired the viability of many financial institutions, but also made them dependent on government bail-outs. The problem of persistent industrial loan default has become a most important and serious issue that has attracted the attention of bankers, financial market operators, international lending institutions such as the World Bank and policy makers in developing countries. Bangladesh is chosen as a case study because it is an interesting example of the persistent industrial loan default problem.
40
Chang, Kai-Yuan, and 張凱原. "A Study of Default Risk of Consumer Loan - A Case of Vehicle Loan." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/53173110765968622405.
Повний текст джерела
41
Chuang, Tzu-Meo, and 莊子妙. "The Default Risk Analysis of Personal Fiduciary Loan." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/10827439252827929428.
Повний текст джерелаАнотація:
碩士國立高雄應用科技大學
金融資訊研究所
101
Domestic financial institutions industry in the consumer finance business of personal Fiduciary loans increasingly competitive environment, to improve the operational order for the output to reduce audit working time, and then seek opportunities aging, provides a simple formula or standardized audit model is definitely necessary . And objective scientific assessment methods to make more credit, credit information obtained quickly and overall system management and analysis, supplemented by loans for effective decision-making judgment. Therefore, financial institutions should establish a set of objective and impartial credit assessment mode, thereby assisting auditors approved quickly and rigorously standardized lending cases and achieve job security possessed the goal. The study of a random sample of 2,000 Bank personal Fiduciary loans as the research object, with loan bid for all the information provided by the borrower to Logistic Regression to analyze the differences between credit point and then consider each loan approval Xiang empirical variables, a systematic analysis and discussion, which found that the most affecting credit risk and occurrence of overdue principal factors for the credit sector in policy making reference. The results of this study show that the borrower marital status, occupation type and whether the holding of real estate three variables on micro-credit loans business credit quality is indeed a high level of explanatory power. Financial institution or credit to this competent to assess the credit risk of loans household level, improving loan quality and reduce operating negligence, risk management is also more accurate and effective.
42
Sun, Jui-tai, and 孫瑞黛. "Unsecured loan default model：Structure change after credit crunch？." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/58739308509490036044.
Повний текст джерелаАнотація:
碩士世新大學
財務金融學研究所(含碩專班)
97
The consumer financial markets are rising after the 1997 Asian Financial Crisis, especially the unsecured personal loans grow rapidly due to high user demand. In 1999 Cosmos Bank is the first bank to launch cash card product and make a big boom. The cash card boom attract all the Taiwan banks launch their products and lastly reach the peak in 2003, and finally cause credit crunch (double-card debt storm). After that, due to our government’s strictly financial supervision and the banks’ self-disciplined, credit crunch are temporarily settled down. But how to control the incident, has become our government’s and the banks most important lesson. Our research study on a Taiwan bank’s unsecured personal loans which loaned two years before and after 2005 credit crunch. We choose 4 most representative samples of small credit and cash card products to explore the impact of customer default risk factors by SPSS statistical software package. 13 covariates include sex (X1), age (X2), education (X3), marital status (X4), home ownership patterns (X5), residence time (X6 ), seniority (X7), the number of dependents (X8), annual income (X9), pieces of credit card (X10), number of recent bank queries (X11), other bank loans (X12), the total number of bank loans (X13) are analyzed for Logistic Regression analysis. From the study analysis result, we find that the 3 common significant covariates of all 4 samples are education (X3), pieces of credit card (X10), number of recent bank queries (X11). We also found that the significant effects of variables of basic customer attributes (X1 ~ X8) are the same. About the cash card sample before credit crunch, all the variables except education (X3) are insignificant, which is different from small credit, but the cash card sample’s trend go in line with the small credit after the credit crunch.
43
Li, Kuei-Jung, and 李桂榮. "Default factors of the owning home mortgage loan." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/60696196361768570042.
Повний текст джерелаАнотація:
碩士國立高雄第一科技大學
金融營運所
91
The past research on mortgage loan default did not separate the mortgages of owning and invested properties. In fact, the purposes and financial burden of these two types of properties are different. The owning properties are used by the owners and are more willing to pay the mortgages. This research only studied the factors that affect the owning home mortgage loan. Data of this research are taken from a branch of bank in southern Taiwan. From the year of 1997 to 2001, in the 426 mortgage loans 136 are default loans and 290 are not default loans. This research used the Logistic regression model to analyze the default factors. It is found that in the eighteen factors, education, housing condition, interest rate, professional, and guarantor have significant impact on default. The higher the education background the less the default rate. New houses have lower default rate. The higher the interest rate the higher the default rate. In terms of professional position, government and school employee have lower default rate. Managers have lower default rate. Those who have guarantors have lower default rate.
44
CHUNG, TUAN HUA, and 段華忠. "Default factors of the investing home mortgage loan." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/84963501014972740646.
Повний текст джерелаАнотація:
碩士國立臺北大學
企業管理學系
93
The past research on mortgage loan default did not separate the mortgages of owning and invested properties. In fact, the purposes and financial burden of these two types of properties are different. The owning properties are used by the owners and are more willing to pay the mortgages. This research only studied the factors that affect the investing home mortgage loan. Data of this research are taken from a branch of bank in southern Taiwan. From the year of 1999 to 2003, in the 500 mortgage loans 50 are default loans and 450 are not default loans. This research used the Logistic regression model to analyze the default factors. It is found that in the eighteen factors, education, housing condition, interest rate, professional, and guarantor have significant impact on default. The higher the education background the less the default rate. New houses have lower default rate. The higher the interest rate the higher the default rate. In terms of professional position, government and school employee have lower default rate. Managers have lower default rate. Those who have guarantors have lower default rate.
45
Hao, Duong Ngoc, and 陽玉豪. "A STUDY OF DEFAULT FACTORS ON PERSONAL LOAN." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/10302565742315587705.
Повний текст джерелаАнотація:
碩士樹德科技大學
金融與風險管理系碩士班
99
In the world, the large bank with more experience, much capital that can be threated to go to bankruptcy because of low credit quality. That reasons are valuable studies for Vietnam banking system. In Viet Nam, 60-70 percentage profits of Vietnam banking system are collected from credit income results. So credit risk is higher than others. Improving credit quality, risk reduction, it’s imperative for them. Although, there are some reports on the credit quality, my thesis is focused on default factors on personal loan. The study applies the method of non-probability sampling, specifically convenience sampling method. Linear regression is used to answer research questions. Result shows that marital status, occupation, age, education, gender and income are statistically significant with significance level of 5%. Thus marital status, occupation, age, education, gender and income are related to the credit quality. In addition, the reasons that make personal loans overdue were presented.
46
Huang, Yu-Hsiang, and 黃郁翔. "A Study in Default Factors of Auto Loan." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/s98xzv.
Повний текст джерелаАнотація:
碩士國立臺灣大學
經濟學研究所
107
A decade after the U.S. subprime crisis in 2007-08, much ink has been spilled on the risks of U.S. subprime lending again. However, the subjects has been switched from house to automobile. In 2016, the subprime auto loans has reached US260 billions, surpassing the level before the global financial crisis. While in Taiwan, the auto sales, as well as auto loan’s share in consumer finance, has been increased year by year since 2009. A vehicle, as opposed to a house, loses its value over time. Therefore, the purpose of this study was to discuss the risk management in financial institutions and to address its influence on expected income. We collected 37,534 auto loans from a domestic financial institution as research samples and employed Logistic regression model for testing the significance of variables. From the empirical results, the loan characteristics with forecasting power include term, loan-to-value ratios, rate, educational level of borrowing applicants, seniority, income and guarantor. The loan interest rate and the delinquency and default rates on auto financing loans are negatively correlated, distinct from that on new and used car loans. The difference may arise from the borrowing purpose, as most subprime borrowers tend to have financing needs rather than those who borrow to buy a car.
47
Rodrigues, Felipe Augusto Silva. "Determinants of Loan Default in Peer-to-Peer Lending at Different Loan Risk Levels." Master's thesis, 2020. https://hdl.handle.net/10216/139323.
Повний текст джерела
48
Rodrigues, Felipe Augusto Silva. "Determinants of Loan Default in Peer-to-Peer Lending at Different Loan Risk Levels." Dissertação, 2020. https://hdl.handle.net/10216/133139.
Повний текст джерела
49
Tsai, Wei-jen, and 蔡偉仁. "A Study on Loss Given Default for Corporate Loan." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/q382jd.
Повний текст джерелаАнотація:
碩士東吳大學
國際經營與貿易學系
96
This study is extracted from a local commercial bank that has sold non-performing loans to the asset management company by choosing 135 cases of corporate loans from 1992 to 2002 as samples. In this article the main direction of this research is to discuss on the influence of the variable factors with LGD (loss given default) and the construction of a LGD model, so as to reduce the non-performing loans of a bank and to improve its competitiveness. First of all, we use the multiple linear regression analysis to forecast the variable factors and to discuss which one can influence the LGD. The experimental evidence showed that the enterprise period of service, loan rate, mortgage, house or stock take for collateral, the economical growth rate and the unemployment rate and the remaining seven variables, have a remarkable influence on the LGD. Secondly, we attempt to construct a LGD estimation model which can discriminate the LGD based on different characteristics of the economy. After this demonstration, its calculation and the prime number only differ by 3.78%, whereby the error is not big. Therefore this model is useful for the LGD examination when the lender applies for loan from financial institution.
50
Hua, Hsueh Cheng, and 華學誠. "The Analysis of the Default Factors on Mortgage Loan." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/22196240783106004869.
Повний текст джерелаАнотація:
碩士世新大學
經濟學研究所(含碩專班)
98
Mortgage Loan is one of important services in the bank, in which housing loans have been accounted for a great portion. However the regulation relaxed and allowing the new banks to launch in Taiwan, the competition among banks became competitive. In order to pursue the higher market share the banks adopted the loosened credit policy or lower price policy to promote products and attract customers. The competition result in the worse credit quality and the more serious overdue loans. Besides, the current economic downturn affected the financial market. In order to analyze the default factors on mortgage loan, the study selects 382 cases of mortgage loan in a domestic commercial bank from 2003 to 2007 after deducting the loan settled. The study applies the Logistic Regression model to carry out the analysis and the discussion of the default factors on the mortgage loan. The findings show that, education and loan interest rate are two explanatory variables which affect borrowers’ paying behavior significantly. In which loan interest rate shows a positive correlation with paying normally, while the school record shows a inverse correlation. All of two are conformed with the anticipated supposition.