karagul Goto Github PK

credit-risk-management-project

towards a use of copula based methods in credit risk management

credit-risk-model

Credit Risk Modelling for Loan Data

credit-risk-model-logistic-regression

Step by step process of building a credit risk model using logistic regression algorithm and also computing EAD (Exposure at default) and LGD (Loss given default)

credit-risk-modeling

Estimated Expected Loss by modelling PD, LGD and EAD and created a scorecard to identify customers that can default on a given consumer loan in future and estimated capital requirement (Capital Adequacy or regulatory capital) using Basel II and Basel III regulations. Used consumer loan dataset on Kaggle from year 2007-2015 issued by LendingClub company. Used pre-processing techniques such as Weight of Evidence, Fine classing and Coarse Classing before building a Probability Density model. Received an AUC score of 73%.

credit-risk-modeling-1

All the way from making a PD_model --> score card ---> calculating LGD, EAD and the Expected Loss

credit-risk-modeling-2

Credit Risk Models for Scorecards, PD, LGD, EAD

credit-risk-modeling-3

This project uses Python and its libraries for implementation of PD (Probability of Default),LGD(Loss Given Default) and EAD(Exposure At Default) models. The models are trained on a dataset, consisting of more than 800,000 consumer loans issued from 2007 to 2014 by Lending Club: a large US peer-to-peer lending company. The Credit Score is also calculated based on demographics (e.g., age, time at residence, time at job, postal code), existing relationship (e.g., time at bank, number of products, payment performance, previous claims), credit bureau (e.g., inquiries, trades, delinquency, public records), real estate data, and so forth. Also the approval and rejection rates are estimated for each possible threshold level and Credit Score ,based on which cut-off value can be set by banks. Finally the Expected Loss is calculated based on the entire portfolio.

credit-risk-modeling-4

Credit Risk Modeling by creating Probability of Default (PD), Loss Given Default (LGD) & Exposure At Default (EAD) models for Credit Risk Management.

credit-risk-modeling-pd-lgd-and-ead-model

Credit Risk Model that predicts the Probability of Default, Loss Given Default and Exposure at Default using Python.

credit-risk-modeling-using-machine-learning

The project goal is to predict expected loss (PD*LGD*EAD) and to generate score card using machine learning so that lenders can make a decision whether to grant a loan or not.

credit-risk-modelling

Credit-Risk Modelling Libraries

credit-risk-modelling-1

Modelling on PD,LGD,EAD and scorecard

credit-risk-modelling-2

With the use of Machine Learning Algorithms, the implementation of the metioned 3 models of PD, LGD and EAD uses various classifier algorithms for PD such as: Logistic Regression(Binary), Random Forest Classifier, Gradient Boost Classifier and Adaboost along with Logistic and Linear Regression Algorithms for evaluating LGD and EAD models.

credit-risk-modelling-3

Predicting weather a loan would default or not. Its more than just a classification problem. Generating scorecards,calculating expected loss, probability of default, LGD, EAD for each respective applicant.

credit-risk-modelling-4

Finding Probability of Default, Loss Given Default and Exposure at Default to find the Expected Loss according to Basel II norms

credit-risk-modelling-lgd-beta-regression-implementation

credit-risk-modelling-social-lending

In this repository, I build interpretable machine learning models to develop credit scorecards by modelling all three components of expected credit loss (probability of default (PD), loss given default (LGD) and exposure at default (ED)

credit-risk-modelling-using-deep-learning

credit-risk-models

Building Base II, Base III compliant credit risk models in Python (PD, LGD, EAD)

credit-risk-models-pd-lgd-ead-expected-loss

We calculate PD,LGD,EAD and Expected loss using logistic and beta regression.

credit-risk-predicting-shiny-dashboard

credit-risk-prediction-and-boston-housing-price-prediction

The aim of this project is to predict credit risk defaulters and housing prices in Boston

credit-risk-scorecard_rmarkdown

credit-risk-var-cvar

credit-risk-vintage-analysis

credit-risk.machine-learning-application.

This work aim to, on one side, the backend code develop on Python Notebook (code_credit_scoring_np.ipynb), for reason of readability, to find the best Hyperparameter in order to achieve the best forecast for the credit scoring with algorithm we analize during the thesis

credit-spreads

credit-suisse-ml-hackathon

Explanatory Stock Price Predictions

creditanalytics-loan-prediction

A predictive model that uses several machine learning algorithms to predict the eligibility of loan applicants based on several factors

creditdefaultswappricer

Credit Default Swap Pricer