Academic literature on the topic 'Default Probability Prediction'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Default Probability Prediction.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Journal articles on the topic "Default Probability Prediction"
1
Tserng, H. Ping, Po-Cheng Chen, Wen-Haw Huang, Man Cheng Lei, and Quang Hung Tran. "PREDICTION OF DEFAULT PROBABILITY FOR CONSTRUCTION FIRMS USING THE LOGIT MODEL." JOURNAL OF CIVIL ENGINEERING AND MANAGEMENT 20, no. 2 (2014): 247–55. http://dx.doi.org/10.3846/13923730.2013.801886.
Full textAbstract:
Recently, the high incidence of construction firm bankruptcies has underlined the importance of forecasting defaults in the construction industry. Early warning systems need to be developed to prevent or avert contractor default; additionally, this evaluation result could facilitate the selection of firms as collaboration or investment partners. Financial statements are considered one of the key basic evaluation tools for demonstrating firm strength. This investigation provides a framework for assessing the probability of construction contractor default based on financial ratios by using the Logit model. A total of 21 ratios, gathered into five financial groups, are utilized to perform univariate logit analysis and multivariate logit analysis for assessing contractor default probability. The empirical results indicate that using multivariate analysis by adding market factor to the liquidity, leverage, activity and profitability factors can increase the accuracy of default prediction more than using only four financial factors. While considering the market factor in the multivariate Logit model, clear incremental prediction performance appears in 1-year evaluation. This study thus suggests that the market factor comprises important information to increase the prediction performance of the model when applied to construction contractors, particularly in short-term evaluation.
2
Amzile, Karim, and Mohamed Habachi. "Assessment of Support Vector Machine performance for default prediction and credit rating." Banks and Bank Systems 17, no. 1 (2022): 161–75. http://dx.doi.org/10.21511/bbs.17(1).2022.14.
Full textAbstract:
Predicting the creditworthiness of bank customers is a major concern for banking institutions, as modeling the probability of default is a key focus of the Basel regulations. Practitioners propose different default modeling techniques such as linear discriminant analysis, logistic regression, Bayesian approach, and artificial intelligence techniques. The performance of the default prediction is evaluated by the Receiver Operating Characteristic (ROC) curve using three types of kernels, namely, the polynomial kernel, the linear kernel and the Gaussian kernel. To justify the performance of the model, the study compares the prediction of default by the support vector with the logistic regression using data from a portfolio of particular bank customers. The results of this study showed that the model based on the Support Vector Machine approach with the Radial Basis Function kernel, performs better in prediction, compared to the logistic regression model, with a value of the ROC curve equal to 98%, against 71.7% for the logistic regression model. Also, this paper presents the conception of a support vector machine-based rating tool designed to classify bank customers and determine their probability of default. This probability has been computed empirically and represents the proportion of defaulting customers in each class.
3
Zhang, Zixuan. "Credit Card Default Prediction based on Machine Learning Techniques." BCP Business & Management 44 (April 27, 2023): 779–85. http://dx.doi.org/10.54691/bcpbm.v44i.4954.
Full textAbstract:
In recent years, with the development of society and economy, credit cards have been popularized due to their low interest rate and easy payment. However, with the advent of the epidemic era, the unemployment rate has increased, making the probability of credit card defaults rising. The prediction of credit card default helps banks and financial institutions balance the risk and economic interests, contributes to the stable and healthy development of the financial industry, and plays an important role in bank credit control. Therefore, this paper addresses the credit card default prediction problem by using Random forest, Decision tree, LightGBM, XGBoost, Logistic regression, and Adaboost models to make predictions and compare the results. The outcomes demonstrate that LightGBM algorithm has the most outstanding prediction score, and its AUC value can reach 0.78 and recall rate reaches 0.95.
4
Lux, Nicole, and Sotiris Tsolacos. "Loan Characteristics as Predictors of Default in Commercial Mortgage Portfolios." International Journal of Economics and Financial Research, no. 71 (February 17, 2021): 1–4. http://dx.doi.org/10.32861/ijefr.71.1.4.
Full textAbstract:
This paper examines the role of loan characteristics in mortgage default probability for different mortgage lenders in the UK. The accuracy of default prediction is tested with two statistical methods, a probit model and linear discriminant analysis, using a unique dataset of defaulted commercial loan portfolios provided by sixty-six financial institutions. Both models establish that the attributes of the underlying real estate asset and the lender are significant factors in determining default probability for commercial mortgages. In addition to traditional risk factors such as loan-to-value and debt servicing coverage ratio lenders and regulators should consider loan characteristics to assess more accurately probabilities of default.
5
Ma, Yuhan. "Prediction of Default Probability of Credit-Card Bills." Open Journal of Business and Management 08, no. 01 (2020): 231–44. http://dx.doi.org/10.4236/ojbm.2020.81014.
Full text
6
Pribadi, Firman, and Susanto Susanto. "Merton model as predictor of failure probability of public banks in Indonesia." Journal of Economics, Business & Accountancy Ventura 17, no. 3 (2015): 393. http://dx.doi.org/10.14414/jebav.v17i3.361.
Full textAbstract:
This research attempts to use Black-Schole-Merton (BSM) model based on market approach to predict default probability of publishing bank in Indonesia. This is done by using stock prices and financial report. In this effort, this study estimates the neutral risk and default probability for the publish bank. The result showed that option model can predict default status more with accurate event long before default information was published for public. It can be studied from the case of Bank Century that has been imposed as a failure bank, in which it is known as bailout bank by the Indonesian government. The model does not only provide the ordinal ranking for the bank sample but also the good early warning prediction for the public. The probability estimation based on the option model can be an innovative model to measure and manage credit risk on the future for predicting probability default in Indonesia.
7
Wang, Zhao, Cuiqing Jiang, and Huimin Zhao. "Depicting Risk Profile over Time: A Novel Multiperiod Loan Default Prediction Approach." MIS Quarterly 47, no. 4 (2023): 1455–86. http://dx.doi.org/10.25300/misq/2022/17491.
Full textAbstract:
With the rapid development of fintech, the need for dynamic credit risk evaluation is becoming increasingly important. While previous studies on credit scoring have mostly focused on single-period loan default prediction, we call for a new avenue—multiperiod default prediction (MPDP)—to depict risk profiles over time. To address the challenges raised by MPDP, such as monotonic default probability prediction and complex relationship accommodation, we propose a novel approach, hybrid and collective scoring (HACS). We design a hybrid modeling strategy to predict whether and when a borrower will default separately through a default discrimination model and a default time estimation model, respectively, and synthesize them through a probabilistic framework. To accommodate various possible patterns of default time and measure the distribution of default probability over successive time intervals, we propose a joint default modeling method to train the default time estimation model. Empirical evaluations at the model (time-to-default prediction performance and discrimination performance) and mechanism (identifiability and discriminability) levels, as well as impact analyses at the application (granting performance and profitability performance) level, show that HACS outperforms the benchmarked survival analysis and multilabel learning methods on all fronts. It can more accurately predict time-to-default and provide financial institutions and investors better decision-support in granting loans and selecting loan portfolios.
8
Matenda, Frank Ranganai, and Mabutho Sibanda. "Determinants of Default Probability for Audited and Unaudited SMEs Under Stressed Conditions in Zimbabwe." Economies 10, no. 11 (2022): 274. http://dx.doi.org/10.3390/economies10110274.
Full textAbstract:
Using stepwise logistic regression models, the study aims to separately detect and explain the determinants of default probability for unaudited and audited small-to-medium enterprises (SMEs) under stressed conditions in Zimbabwe. For effectiveness purposes, we use two separate datasets for unaudited and audited SMEs from an anonymous Zimbabwean commercial bank. The results of the paper indicate that the determinants of default probability for unaudited and audited SMEs are not identical. These determinants include financial ratios, firm and loan characteristics, and macroeconomic variables. Furthermore, we discover that the classification rates of SME default prediction models are enhanced by fusing financial ratios and firm and loan features with macroeconomic factors. The study highlights the vital contribution of macroeconomic factors in the prediction of SME default probability. We recommend that financial institutions model separately the default probability for audited and unaudited SMEs. Further, it is recommended that financial institutions should combine financial ratios and firm and loan characteristics with macroeconomic variables when designing default probability models for SMEs in order to augment their classification rates.
9
Sharma, Meena, and Sakshi Khurana. "Impact of Effective Generic Business Strategies on Probability of Default: A Study of NSE Listed Firms." NMIMS Management Review 33, no. 1 (2025): 29–39. https://doi.org/10.1177/09711023251323380.
Full textAbstract:
The aim of the current study is to examine the impact of generic business strategies (GBS) on default risk for Indian firms listed on the National Stock Exchange. This study employs a distance-to-default model to compute default risk and GBS, which are measured through Porter’s framework. The relationship between GBS and the probability of default is derived through panel data regression analysis for the period 2008–2009 to 2021–2022. The results support the hypothesis and show a positive relationship between cost leadership and the distance-to-default model. The findings indicate that companies employing a cost leadership strategy tend to have sound financial health, which makes them less prone to default risk. The results show practical implications for financial institutions, creditors, and investors who can check the strategic positioning of a firm before evaluating and arriving at the credit rating, lending, and investment decisions. Based on the findings of this study, investors can provide funds to invest in a firm that implements GBS effectively to ensure high investment returns and fewer credit defaults. The study is expected to contribute to default risk literature by providing evidence that GBS indicators can be experimented for incorporation into new default risk prediction models.
10
GRIGOROVICH, Anastasiya V., Dmitrii V. GRIGOROVICH, and Vadim M. KOZHANOV. "Logit model for predicting the probability of default for Russian small and medium-sized enterprises." Finance and Credit 30, no. 2 (2024): 390–417. http://dx.doi.org/10.24891/fc.30.2.390.
Full textAbstract:
Subject. This article deals with the issues of determining the financial stability of small and medium-sized enterprises (SMEs) in terms of the probability of future default. Objectives. The article aims to determine the most accurate model for predicting the probability of default of Russian SMEs, as well as develop a new model based on an up-to-date database of financial indicators obtained from open sources of information. Methods. For the study, we used logistic regression and analysis. Results. The article presents a developed six-factor logit model for estimating the probability of default a year before its occurrence, which has shown the highest accuracy of prediction in comparison with the models available in the literature. Relevance. The developed logit model may be of interest to a wide range of users wishing to invest in Russian SMEs due to its wide analytical base, high prediction accuracy, and ease of use and interpretation of results.
Dissertations / Theses on the topic "Default Probability Prediction"
1
Petrosyan, Mariam <1990>. "The role of volatility persistence in default probability prediction: A Bayesian model." Master's Degree Thesis, Università Ca' Foscari Venezia, 2017. http://hdl.handle.net/10579/10800.
Full textAbstract:
This thesis studies the significance of lagged value of stochastic volatility process in the prediction of probability of default in one year, and in estimation of credit default swap spreads. The novelty of the model consists in allowing for both stochastic interest rate and volatility and in the extension of the dynamics of SV process to AR (2) process. The estimation is carried out by exploiting Bayesian methods via implementation of Gibbs sampling for the state space model of stochastic volatility and returns and Metropolis Hastings acceptance rejection sampling algorithms. The model is compared with Merton’s basic SCR model and with the SCR model with stochastic interest rate presented in Rodriguez et al (2014) on the basis of marginal likelihoods, using financial data series of three firms .
2
Mirzaei, Maryam [Verfasser]. "Corporate Probability Default Prediction With Industry Effects Using Data Mining Techniques / Maryam Mirzaei." Munich : GRIN Verlag, 2016. http://d-nb.info/1114737127/34.
Full text
3
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.
Full textAbstract:
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.<br>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.
4
Caetano, João Manuel Nunes. "Predictive models of probability of default : an empirical application." Master's thesis, Instituto Superior de Economia e Gestão, 2014. http://hdl.handle.net/10400.5/7704.
Full textAbstract:
Mestrado em Finanças<br>Este estudo tem como objetivo realizar uma pesquisa dos modelos de previsão do incumprimento a empresas listadas em bolsa. Foram abordadas as metodologias do modelo de Merton (1974), modelo Contabilístico e Híbrido. Testou-se uma amostra de 172 empresas presentes no mercado Americano dos setores do Consumo, Distribuição, Produção e Telecomunicações nas quais 82 entram em incumprimento. Para cada metodologia, a capacidade preditiva foi testada através dos erros Tipo I e II. Os resultados sugerem que o modelo Híbrido, i.e., a combinação de modelos de mercado e análise contabilística, confere maior poder de precisão na classificação de incumprimento, ao invés de cada modelo individualmente.<br>This study intends to conduct a survey of Probability of Default models to listed companies. The methodologies of Merton (1974) model, Accounting model and Hybrid were addressed. We tested a sample of 172 American companies in the sectors of Consumer Products, Distribution, Manufacturing and Telecommunications in which 82 entered into default. For each methodology, the predictive ability was tested with Type I and II errors. The results suggests that the Hybrid model, i.e. a combination of market models and accounting analysis, have a better performance in the classification of credit default than each model individually.
5
Kornfeld, Sarah. "Predicting Default Probability in Credit Risk using Machine Learning Algorithms." Thesis, KTH, Matematisk statistik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-275656.
Full textAbstract:
This thesis has explored the field of internally developed models for measuring the probability of default (PD) in credit risk. As regulators put restrictions on modelling practices and inhibit the advance of risk measurement, the fields of data science and machine learning are advancing. The tradeoff between stricter regulation on internally developed models and the advancement of data analytics was investigated by comparing model performance of the benchmark method Logistic Regression for estimating PD with the machine learning methods Decision Trees, Random Forest, Gradient Boosting and Artificial Neural Networks (ANN). The data was supplied by SEB and contained 45 variables and 24 635 samples. As the machine learning techniques become increasingly complex to favour enhanced performance, it is often at the expense of the interpretability of the model. An exploratory analysis was therefore made with the objective of measuring variable importance in the machine learning techniques. The findings from the exploratory analysis will be compared to the results from benchmark methods that exist for measuring variable importance. The results of this study shows that logistic regression outperformed the machine learning techniques based on the model performance measure AUC with a score of 0.906. The findings from the exploratory analysis did increase the interpretability of the machine learning techniques and were validated by the results from the benchmark methods.<br>Denna uppsats har undersökt internt utvecklade modeller för att estimera sannolikheten för utebliven betalning (PD) inom kreditrisk. Samtidigt som nya regelverk sätter restriktioner på metoder för modellering av kreditrisk och i viss mån hämmar utvecklingen av riskmätning, utvecklas samtidigt mer avancerade metoder inom maskinlärning för riskmätning. Således har avvägningen mellan strängare regelverk av internt utvecklade modeller och framsteg i dataanalys undersökts genom jämförelse av modellprestanda för referens metoden logistisk regression för uppskattning av PD med maskininlärningsteknikerna beslutsträd, Random Forest, Gradient Boosting och artificiella neurala nätverk (ANN). Dataunderlaget kommer från SEB och består utav 45 variabler och 24 635 observationer. När maskininlärningsteknikerna blir mer komplexa för att gynna förbättrad prestanda är det ofta på bekostnad av modellens tolkbarhet. En undersökande analys gjordes därför med målet att mäta förklarningsvariablers betydelse i maskininlärningsteknikerna. Resultaten från den undersökande analysen kommer att jämföras med resultat från etablerade metoder som mäter variabelsignifikans. Resultatet av studien visar att den logistiska regressionen presterade bättre än maskininlärningsteknikerna baserat på prestandamåttet AUC som mätte 0.906. Resultatet from den undersökande analysen för förklarningsvariablers betydelse ökade tolkbarheten för maskininlärningsteknikerna. Resultatet blev även validerat med utkomsten av de etablerade metoderna för att mäta variabelsignifikans.
6
Pörn, Sebastian, and Arvid Rönnblom. "Assesing counterparty risk classification using transition matrices : Comparing models' predictive ability." Thesis, Umeå universitet, Institutionen för matematik och matematisk statistik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-136667.
Full textAbstract:
An important part when managing credit risk is to assess the probability of default of different counterparties. Increases and decreases in such probabil- ities are central components in the assessment, and this is where transition matrices become useful. These matrices are commonly used tools when as- sessing counterparty credit risk, and contain the probability of default, as well as the probability to migrate between different predefined rating classifica- tions. These rating classifications are used to reflect the risk taken towards different counterparties. Therefore, it is important for financial institutions to develop accurate transition matrix models to manage predicted changes in credit risk exposure. This is because counterparty creditworthiness and prob- ability of default indirectly affect expected loss and the capital requirement of held capital. This thesis will analyze how two specific models perform when used for generating transition matrices. These models will be tested to investigate their performance when predicting rating transitions, including probability of default.<br>En viktig del vid hanteringen av kreditrisk är att bedöma sannolikheten för fallissemang för olika motparter. Ökningar och minskningar i dessa sanno- likheter är centrala komponenter i bedömningen, och det är här migrations- matriser blir användbara. Dessa matriser är vanligt förekommande verktyg vid bedömning av kreditrisk mot olika motparter och innehåller sannolikheten för fallissemang samt sannolikheten att migrera mellan olika fördefinierade be- tygsklassificeringar. Dessa betygsklassificeringar används för att återspegla den risk som tas mot olika motparter. Det är därför viktigt för finansinstitut att utveckla träffsäkra migrationsmatris modeller för att hantera förväntade förändringar i kreditriskexponering. Detta beror på att kreditvärdigheten hos motparter samt sannolikheten för fallissemang indirekt påverkar expected loss och kapitalkrav. Detta examensarbete kommer att analysera hur två specifika modeller presterar när de används för att generera migrationsmatriser. Dessa mod- eller kommer att testas för att undersöka hur de presterar när de används för att förutsäga övergångar inom betygsklassificering, inklusive sannolikheten för fallissemang.
7
Lundström, Love, and Oscar Öhman. "Machine Learning in credit risk : Evaluation of supervised machine learning models predicting credit risk in the financial sector." Thesis, Umeå universitet, Institutionen för matematik och matematisk statistik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-164101.
Full textAbstract:
When banks lend money to another party they face a risk that the borrower will not fulfill its obligation towards the bank. This risk is called credit risk and it’s the largest risk banks faces. According to the Basel accord banks need to have a certain amount of capital requirements to protect themselves towards future financial crisis. This amount is calculated for each loan with an attached risk-weighted asset, RWA. The main parameters in RWA is probability of default and loss given default. Banks are today allowed to use their own internal models to calculate these parameters. Thus hold capital with no gained interest is a great cost, banks seek to find tools to better predict probability of default to lower the capital requirement. Machine learning and supervised algorithms such as Logistic regression, Neural network, Decision tree and Random Forest can be used to decide credit risk. By training algorithms on historical data with known results the parameter probability of default (PD) can be determined with a higher certainty degree compared to traditional models, leading to a lower capital requirement. On the given data set in this article Logistic regression seems to be the algorithm with highest accuracy of classifying customer into right category. However, it classifies a lot of people as false positive meaning the model thinks a customer will honour its obligation but in fact the customer defaults. Doing this comes with a great cost for the banks. Through implementing a cost function to minimize this error, we found that the Neural network has the lowest false positive rate and will therefore be the model that is best suited for this specific classification task.<br>När banker lånar ut pengar till en annan part uppstår en risk i att låntagaren inte uppfyller sitt antagande mot banken. Denna risk kallas för kredit risk och är den största risken en bank står inför. Enligt Basel föreskrifterna måste en bank avsätta en viss summa kapital för varje lån de ger ut för att på så sätt skydda sig emot framtida finansiella kriser. Denna summa beräknas fram utifrån varje enskilt lån med tillhörande risk-vikt, RWA. De huvudsakliga parametrarna i RWA är sannolikheten att en kund ej kan betala tillbaka lånet samt summan som banken då förlorar. Idag kan banker använda sig av interna modeller för att estimera dessa parametrar. Då bundet kapital medför stora kostnader för banker, försöker de sträva efter att hitta bättre verktyg för att uppskatta sannolikheten att en kund fallerar för att på så sätt minska deras kapitalkrav. Därför har nu banker börjat titta på möjligheten att använda sig av maskininlärningsalgoritmer för att estimera dessa parametrar. Maskininlärningsalgoritmer såsom Logistisk regression, Neurala nätverk, Beslutsträd och Random forest, kan användas för att bestämma kreditrisk. Genom att träna algoritmer på historisk data med kända resultat kan parametern, chansen att en kund ej betalar tillbaka lånet (PD), bestämmas med en högre säkerhet än traditionella metoder. På den givna datan som denna uppsats bygger på visar det sig att Logistisk regression är den algoritm med högst träffsäkerhet att klassificera en kund till rätt kategori. Däremot klassifiserar denna algoritm många kunder som falsk positiv vilket betyder att den predikterar att många kunder kommer betala tillbaka sina lån men i själva verket inte betalar tillbaka lånet. Att göra detta medför en stor kostnad för bankerna. Genom att istället utvärdera modellerna med hjälp av att införa en kostnadsfunktion för att minska detta fel finner vi att Neurala nätverk har den lägsta falsk positiv ration och kommer därmed vara den model som är bäst lämpad att utföra just denna specifika klassifierings uppgift.
8
Fang, Ming-Ching, and 房明慶. "The Prediction of probability of default in the Public company." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/2jcz3r.
Full textAbstract:
碩士<br>東吳大學<br>商用數學系<br>96<br>Credit risk management has become the main requirement of Basel II Capital regulatory. Then we used credit risk tools, the most important thing is credit rating which was derived from the information of counterparty repayment ability. Thus we give a compatible rating from evaluation result for each borrower. First of all, the new accounts information of enterprises and variables of financial statement were employed by building of credit rating model, and in accordance of above-mentioned information to predict a probability of default. On a conservative basis, we used logistic regression methodology to evaluate the 1097 public companies, and exercised 4959 samples data between 87 and 93 years. Also, we derived the development model samples from the 5 years between 87 and 91 years, validation model samples from 2 years in 92 and 93 years. Financial variables analysis was included by financial construction, liquidity, cash flow, operation efficiency, and profit ability. We selected the 18 variables from all of 100 variables by AUC>65% and logistic regression rules. Also, final 6 variables were used by binning methodology. The AUC value of discriminatory development model is 87.1%, and validation model was used the final coefficient and variables from development model. Also, AUC is 81.3% to separate the good or bad enterprises. The final result was effected by outstanding variables of the public company were profitability、sales revenue/liquidity debt、 stakeholders sales、 loan debt and debt ratio. Finally, we used the methodology of cluster analysis to separate ratings by predicted PD, and followed Basel II rules that each samples of ratings can not be over than in 30% and minimum 7 ranking. In order to avoid samples were overly centralized. Finally, the ratings distribution of public company is right trend instead of normal distribution of consumer baking.
9
Chu, Tzu-Lin, and 朱賜麟. "Effectiveness of the Default-Corpus from Linguistic Data Mining on the Prediction of Corporate Default Probability." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/r6rv3g.
Full textAbstract:
碩士<br>銘傳大學<br>財務金融學系碩士班<br>97<br>This paper applies the linguistic data mining technique to extract relevant information from financial news. The Chinese corpuses for financial crisis and distress abstracted from our mining process are quantified through text corpus analysis, keywords frequency analysis, entropy estimation and default intensity analysis. The Chinese corpuses for financial crisis and distress are then transformed into a measure concerning the intensity to financial distress. We put such a distress-corpus variable (intensity of default-corpus, ITDC) together with other variables about financial structure, corporation governance, treatment variable, and macroeconomic variables into logistic regression to investigate whether news plays an important role in improving the financial warning capability. The empirical results show that ITDC improve the explanation power of each financial distress model regardless of what independent variables are incorporated. For the prediction effectiveness, the results show that ITDC significantly contribute to reduce the type I error and improve identification accuracy of financial distress. The one-quarter ahead forecast presents the minimum type I error (13.33%) and largest identification accuracy (89.35%). Our results prove that intensity of default-corpus variable from linguistic data mining of Chinese financial news does improve the effectiveness of the prediction of corporate default probability.
10
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.
Full textAbstract:
碩士<br>輔仁大學<br>資訊管理學系碩士班<br>107<br>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.
Book chapters on the topic "Default Probability Prediction"
1
Łuczak, Aleksandra, Maria Ganzha, and Marcin Paprzycki. "Probability of Loan Default—Applying Data Analytics to Financial Credit Risk Prediction." In Intelligent Systems, Technologies and Applications. Springer Singapore, 2021. http://dx.doi.org/10.1007/978-981-16-0730-1_1.
Full text
2
Zhao, Yi, Yong Huang, and Yanyan Shen. "$$DMDP^2$$DMDP2: A Dynamic Multi-source Based Default Probability Prediction Framework." In Web and Big Data. Springer International Publishing, 2018. http://dx.doi.org/10.1007/978-3-319-96890-2_26.
Full text
3
Martinelli, Fabio, Francesco Mercaldo, Domenico Raucci, and Antonella Santone. "Predicting Probability of Default Under IFRS 9 Through Data Mining Techniques." In Advances in Intelligent Systems and Computing. Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-44038-1_87.
Full text
4
Suárez-Ramírez, Cuauhtémoc Daniel, Juan-Carlos Martínez, and Octavio Loyola-González. "A Novel Survival Analysis-Based Approach for Predicting Behavioral Probability of Default." In Lecture Notes in Computer Science. Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-07750-0_6.
Full text
5
Oleynik, Mariya, and Tomáš Formánek. "Predicting Default Probability of Bank’s Corporate Clients in the Czech Republic. Comparison of Generalized Additive Models and Support Vector Machine Approaches." In Software Engineering Perspectives in Intelligent Systems. Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-63322-6_60.
Full text
6
Lisboa, Ines, and Magali Costa. "International Effect on Family SME Financial Distress Prediction." In Research Anthology on Strategies for Maintaining Successful Family Firms. IGI Global, 2022. http://dx.doi.org/10.4018/978-1-6684-3550-2.ch009.
Full textAbstract:
Understanding the reasons of default risk is crucial to avoid the firm's bankruptcy. The purpose of this work is to analyze the impact of internationalization on firm's probability of distress. For it, this chapter aims to propose a model to predict default specific to family SMEs (small and medium enterprises). An unbalanced panel of 10,832 firms over the period from 2012-2018 is analyzed. Ex-ante criteria to classify firms in default or compliant is used. International SMEs have lower probability of default than domestic firms, and compliant firms export more. Results show that export ratio is an important determinant of the probability of default. Moreover, the ratios of liquidity, profitability, size, leverage, efficiency, cash flow, and age are also relevant. Moreover, these ratios explain default risk of both groups international and domestic SMEs. The proposed model has an accuracy of 92.9%, which increases to 95.6% if only export SMEs are analyzed.
7
Lisboa, Ines, and Magali Costa. "International Effect on Family SME Financial Distress Prediction." In Cases on Internationalization Challenges for SMEs. IGI Global, 2021. http://dx.doi.org/10.4018/978-1-7998-4387-0.ch009.
Full textAbstract:
Understanding the reasons of default risk is crucial to avoid the firm's bankruptcy. The purpose of this work is to analyze the impact of internationalization on firm's probability of distress. For it, this chapter aims to propose a model to predict default specific to family SMEs (small and medium enterprises). An unbalanced panel of 10,832 firms over the period from 2012-2018 is analyzed. Ex-ante criteria to classify firms in default or compliant is used. International SMEs have lower probability of default than domestic firms, and compliant firms export more. Results show that export ratio is an important determinant of the probability of default. Moreover, the ratios of liquidity, profitability, size, leverage, efficiency, cash flow, and age are also relevant. Moreover, these ratios explain default risk of both groups international and domestic SMEs. The proposed model has an accuracy of 92.9%, which increases to 95.6% if only export SMEs are analyzed.
8
"Predicting Loan Applicants' Timely Payments." In Decision and Prediction Analysis Powered With Operations Research. IGI Global, 2024. http://dx.doi.org/10.4018/979-8-3693-4179-7.ch005.
Full textAbstract:
This chapter illustrates a prediction of the loan applicants' timely payments with optimization. A neural networks tool is used to predict unknown values of categorical dependent variables from known values of numeric and categorical independent variables. In this model, a neural net learns to predict whether an auto loan applicant will make timely payments, late payments, or default on the loan. The data contains information on applicants who took car loans in the past. The input data of five new applicants is also given. It is supposed that the bank executives want to allocate a certain amount of money in loans to the five applicants to minimize the probability of a default occurring. Therefore, Neural Networks and optimization tools are used to predict the optimal values for the new applicants.
9
Keramati, Abbas, Niloofar Yousefi, and Amin Omidvar. "Default Probability Prediction of Credit Applicants Using a New Fuzzy KNN Method with Optimal Weights." In Advances in Business Information Systems and Analytics. IGI Global, 2015. http://dx.doi.org/10.4018/978-1-4666-7272-7.ch024.
Full textAbstract:
Credit scoring has become a very important issue due to the recent growth of the credit industry. As the first objective, this chapter provides an academic database of literature between and proposes a classification scheme to classify the articles. The second objective of this chapter is to suggest the employing of the Optimally Weighted Fuzzy K-Nearest Neighbor (OWFKNN) algorithm for credit scoring. To show the performance of this method, two real world datasets from UCI database are used. In classification task, the empirical results demonstrate that the OWFKNN outperforms the conventional KNN and fuzzy KNN methods and also other methods. In the predictive accuracy of probability of default, the OWFKNN also show the best performance among the other methods. The results in this chapter suggest that the OWFKNN approach is mostly effective in estimating default probabilities and is a promising method to the fields of classification.
10
Keramati, Abbas, Niloofar Yousefi, and Amin Omidvar. "Default Probability Prediction of Credit Applicants Using a New Fuzzy KNN Method With Optimal Weights." In Intelligent Systems. IGI Global, 2018. http://dx.doi.org/10.4018/978-1-5225-5643-5.ch082.
Full textAbstract:
Credit scoring has become a very important issue due to the recent growth of the credit industry. As the first objective, this chapter provides an academic database of literature between and proposes a classification scheme to classify the articles. The second objective of this chapter is to suggest the employing of the Optimally Weighted Fuzzy K-Nearest Neighbor (OWFKNN) algorithm for credit scoring. To show the performance of this method, two real world datasets from UCI database are used. In classification task, the empirical results demonstrate that the OWFKNN outperforms the conventional KNN and fuzzy KNN methods and also other methods. In the predictive accuracy of probability of default, the OWFKNN also show the best performance among the other methods. The results in this chapter suggest that the OWFKNN approach is mostly effective in estimating default probabilities and is a promising method to the fields of classification.
Conference papers on the topic "Default Probability Prediction"
1
Nicat, Şahin, and Anıl Ferdi Kaya. "Default Prediction and Vehicle Recovery Probability Analysis in Auto Loans - A Segmentation Study." In 2024 9th International Conference on Computer Science and Engineering (UBMK). IEEE, 2024. https://doi.org/10.1109/ubmk63289.2024.10773584.
Full text
2
Hui Zhou, Yi Wang, Wei Wang, Tao Li, and Hong Yang. "Prediction of default probability of clients' electricity charge arrears." In 2008 IEEE International Conference on Service Operations and Logistics, and Informatics. IEEE, 2008. http://dx.doi.org/10.1109/soli.2008.4682972.
Full text
3
Lux, Nicole. "Relevance of loan characteristics in probability of default prediction for commercial mortgage loans." In 26th Annual European Real Estate Society Conference. European Real Estate Society, 2019. http://dx.doi.org/10.15396/eres2019_86.
Full text
4
Bitetto, Alessandro, Stefano Filomeni, and Michele Modina. "Can unlisted firms benefit from market information? A data-driven approach." In CARMA 2022 - 4th International Conference on Advanced Research Methods and Analytics. Universitat Politècnica de València, 2022. http://dx.doi.org/10.4995/carma2022.2022.15045.
Full textAbstract:
We employ a sample of 10,136 Italian micro-, small-, and mid-sized enterprises (MSMEs) that borrow from 113 cooperative banks to examine whether market pricing of public firms adds additional information to accounting measures in predicting default of private firms. Specifically, we first match the asset prices of listed firms following a data-driven clustering by means of Neural Networks Autoencoder so to evaluate the firm-wise probability of default (PD) of MSMEs. Then, we adopt three statistical techniques, namely linear models, multivariate adaptive regression spline, and random forest to assess the performance of the models and to explain the relevance of each predictor. Our results provide novel evidence that market information represents a crucial indicator in predicting corporate default of unlisted firms. Indeed, we show a significant improvement of the model performance, both on class-specific (F1-score for defaulted class) and overall metrics (AUC) when using market information in credit risk assessment, in addition to accounting information. Moreover, by taking advantage of global and local variable importance technique we prove that the increase in performance is effectively attributable to market information, highlighting its relevant effect in predicting corporate default.
5
Enright, Michael P., R. Craig McClung, Wuwei Liang, Yi-Der Lee, Jonathan P. Moody, and Simeon Fitch. "A Tool for Probabilistic Damage Tolerance of Hole Features in Turbine Engine Rotors." In ASME Turbo Expo 2012: Turbine Technical Conference and Exposition. American Society of Mechanical Engineers, 2012. http://dx.doi.org/10.1115/gt2012-69968.
Full textAbstract:
Over the past two decades, the Federal Aviation Administration (FAA) and the aircraft engine industry (organized through the Rotor Integrity Sub-Committee (RISC) of the Aerospace Industries Association) have been developing enhanced life management methods to address the rare but significant threats posed by undetected material or manufacturing anomalies in high-energy rotating components of gas turbine engines. This collaborative effort has led to the release of several FAA advisory circulars providing guidance for the use of probabilistic damage tolerance methods as a supplement to traditional safe-life methods. The most recent such document is Advisory Circular (AC) 33.70-2 on “Damage Tolerance of Hole Features in High-Energy Turbine Rotors.” In parallel with this effort, the FAA has also been funding research and development activities to develop the technology and tools necessary to implement the new methods, including a series of grants led by Southwest Research Institute® (SwRI®). The most significant outcome of these grants is a probabilistic damage tolerance computer code called DARWIN® (Design Assessment of Reliability With INspection). DARWIN integrates finite element models and stress analysis results, fracture mechanics models, material anomaly data, probability of crack detection, and uncertain inspection schedules with a user-friendly graphical user interface (GUI) to determine the probability of fracture of a rotor disk as a function of operating cycles with and without inspection. This paper provides an overview of new DARWIN models and features that directly support implementation of the new AC on hole features. The paper also simultaneously provides an overview of the AC methodology itself. Component geometry and stresses are addressed through an interface with commercial three-dimensional finite element (FE) models, including management of multiple load steps and multiple missions. Calculations of fatigue crack growth (FCG) life employ a unique interface with the FE models, sophisticated new stress intensity factor solutions for typical crack geometries at holes, shakedown modules, a menu of common FCG equations, and algorithms to address the effects of varying temperatures on crack growth rates. The primary random variables are based on the default anomaly distributions and probability-of-detection (POD) curves provided directly in the AC. Fracture risk is computed on a per-feature basis using one of several available computational methods including importance sampling, response surface, and Monte Carlo simulation. The approach is illustrated for risk prediction of a representative gas turbine engine disk. The results can be used to gain a better understanding of the AC and how the problem is solved using the probabilistic damage tolerance framework provided in DARWIN.
6
Barua, Sujoy, Divya Gavandi, Pooja Sangle, Leena Shinde, and Jyoti Ramteke. "Swindle: Predicting the Probability of Loan Defaults using CatBoost Algorithm." In 2021 5th International Conference on Computing Methodologies and Communication (ICCMC). IEEE, 2021. http://dx.doi.org/10.1109/iccmc51019.2021.9418277.
Full text
7
Murcia, Catalina Lozano, and Francisco P. Romero. "Explainability in Models for Predicting the Probability of Default in Compliance with International Standard IFRS 9." In 2021 16th Iberian Conference on Information Systems and Technologies (CISTI). IEEE, 2021. http://dx.doi.org/10.23919/cisti52073.2021.9476455.
Full textReports on the topic "Default Probability Prediction"
1
Tirapat, Sunti. An investigation of default probability in Thailand. Chulalongkorn University, 2001. https://doi.org/10.58837/chula.res.2001.21.
Full textAbstract:
Using the sample of 100 most liquid companies listed in the Stock Exchange of Thailand during 1992-1999, the default probabilities from two approaches, the logit model and the KMV model, are calculated and compared. The results from the KMV model suggest that the default probabilities of financial institutions are higher than the probabilities of industrial companies. Moreover, the results from the KMV model confirm that the average default probabilities of the financial distressed firms in the 1997 financial crisis are higher than the average default probabilities of non-distressed firms. Comparing the prediction of the KMV model with the logit model, the results show that the logit model is better in terms of total prediction error and the Type I error at any cut off levels. The regression results suggest that the default probabilities of the two models have positive associations and seem to be consistent over the period of 1992-1999. Finally, the study examines whether the default probabilities have been priced. The results suggest that investors indeed do require compensations for default risk. The evidence also suggests that investors are more concerned of risk and require higher compensation for likelihood of default after the financial crisis.
2
Blanco, Roberto, Elena Fernández, Miguel García-Posada, and Sergio Mayordomo. An estimation of the default probabilities of Spanish non-financial corporations and their application to evaluate public policies. Banco de España, 2023. http://dx.doi.org/10.53479/33512.
Full textAbstract:
We model the one-year ahead probability for default of Spanish non-financial corporations using data for the period 1996-2019. While most previous literature considers that a firm is in default if it files for bankruptcy, we define default as having non-performing loans during at least three months of a given year. This broader definition allows us to predict firms’ financial distress at an earlier stage that cannot generally be observed by researchers, before their financial conditions become too severe and they have to file for bankruptcy or engage in private workouts with their creditors. We estimate, by means of logistic regressions, both a general model that uses all the firms in the sample and six models for different size-sector combinations. The selected explanatory variables are five accounting ratios, which summarise firms’ creditworthiness, and the growth rate of aggregate credit to non-financial corporations, to take into account the role of credit availability in mitigating the risk of default. Finally, we carry out two applications of our prediction models: we construct credit rating transition matrices and evaluate a programme implemented by the Spanish government to provide direct aid to firms severely affected by the COVID-19 crisis.