Freelancer, Python, Loan Default Model using ensemble methods, Corp of personal loans that matches borrowers who are seeking a loan with investors looking to lend money and make a return. The orp evaluates each borrower’s credit score using past historical data and predicts whether a loan will default or repay the loan., Logistic Regression, Decision trees, Random Forest, Naïve Bayes, SVM, KNN, Cross Validation, Synthetic Oversampling Technique (SMOTE) Freelancer, R, Employee Attrition Prediction using ensemble methods, The factors that cause employees to leave the organization and explore important questions such as ‘compare average monthly income by education and attrition’ or ‘is distance from home a case for attrition’? Build classification models to predict which employee is likely to churn and help the business to devise policies and attract back the right talent., logistic regression and CART First Indian Corporation, SAS and SQL, Model building at securities side, logistic regression, Worked on model building of a product RiskModel which has been significantly used to understand the default and prepayment risk of higher-quality loans, including Agency mortgages. There I was involved in the model building on securities data which needs to be refreshed after every quarter. It was built on basis of Prime, Subprime and ALTA loans. Worked with US team to enhance it as in 2007 the loans were very less at securities side, it needed to be built on the current market scenario. First Indian Corporation, SAS and SQL, Whether servicing data is appropriate for modelling or not?, SAS BASE and logistic regression, Servicing Data is contributed regularly by the leading U.S. servicers on more than 37 million active mortgages, with delinquency, prepayment, and roll-rate metrics —and thousands of analytics options on the full dataset or customer-defined cohorts. Post 2007 the securitization was not happening, but servicers side the loan were originating. I was involved in the exploratory analysis of the servicing data along with 2 more member and was recognized for leading my team of 2 members. Cranes International, SAS, SPSS, SPLUS, SYSTAT, Development of statistical algorithm linear regression in SYSTAT 12.0, a statistical package, SYSTAT, a desktop statistical computing, menu-driven interface to conduct simple analyses and produce beautiful 2D and 3D graphics for presentations or report. Prepared algorithms for linear regression and implemented in SYSTAT 12.0. Implemented confidence intervals, prediction intervals and VIF. Cranes International, SYSTAT, SAS, SPSS, Development of statistical algorithm Best Subset regression in SYSTAT 13, Best subsets regression is an exploratory model building regression analysis. It compares all possible models that can be created based upon an identified set of predictors. Developed the algorithm and compared it with SAS. Cranes International, SYSTAT, SAS, SPSS, Development of statistical algorithm Receiver Operating Characteristic Curve in SYSTAT 13, ROC curve is a performance measurement for classification problem at various thresholds settings. Implemented the algorithm in SYSTAT 13.