Debashis Dutta's repositories
CECL-Modelling-Implementation
Forecasted the Expected Credit Loss, over the lifetime of the mortgage. Built Loan-level PD Model using Markov Chain Transition Matrix and logistic regression with six transition states and validated them using backtesting.
Peer-graded-Assignment-Segmenting-and-Clustering-Neighborhoods-in-Toronto
Peer-graded Assignment: Segmenting and Clustering Neighborhoods in Toronto
Coursera_Capstone-8
Capstone Project Notebook - Week 1
ibm-capstone
Repository for code generated during the IBM/Coursera Data Science Professional capstone project
CreditRiskPaper
Codes for replication and implementation of techniques in our credit risk article
Peer-graded-Assignment-Capstone-Project---The-Battle-of-Neighborhoods-Week-2-
Peer-graded Assignment: Capstone Project - The Battle of Neighborhoods (Week 2)
Applied-Data-Science-Capstone
IBM Data Science Professional Certificate
IBM-DataScienceSpecialization-CAPSTONE
This notebook is for Coursera's Data Science Specialization CAPSTONE
Credit_ML_VaR
Apply a series of machine learning models on financial data to predict credit default risk. Evaluate the credit risk of a portfolio (build by sample data) by VaR and ES models. Validate ML models' performance and check the models' forecast errors in VaR evaluation.
Credit-Risk-Modeling-using-Machine-Learning
A study and comparison of Risk Modeling algorithms (Capstone Project)
DL_forFinance
This git repository is based on the work of J.Heaton, N.Polson and J.Witte and their articleDeep Learning for Finance: Deep Portfolios. This paper let us explore the use of deeplearning models for problems in financial prediction and classification. Our goal isto show how applying deep learning methods to these problems can produce betteroutcomes than standard methods in finance or in Machine Learning
Credit-Risk-Model-1
Read Me File: 1. Type into the command line: cd Desktop 2. Drag the DecisionTree.java file along with all the data file on the Desktop ( Or skip the step 1 and directly drag DecisionTree. java file along with all the data file to the default directory path) 3.Type into the command line: javac DecisionTree.java (After this step , should generate three class file) 4.Type into the command line: java DecisionTree 1 training_set.csv validation_set.csv test_set.csv 1 (Or other argument or data file name that match the input argument format) Random Attribute Selection part: This part in the code starts from line 567. Unlike chooseBestAttribute method which choosing the largest IG of each attribute, chooseRandomtAttribute choose a random attribute column and return to the generate decision tree(line 431) method for next iteration, and we will use random generator to choose the random index and guarantee randomness.To test the decision tree based on random attribute selection, simply replace line 476 "chooseBestAttribute" with "chooseRandomAttribute".
credit-risk
Credit Risk - IRB Model Validation - BASEL Requirement