SUMMARY

• Advance Analytics and FRM professional with 10+ years of experience in delivering Statistical Modeling solutions in Credit risk model validation and Data analytics. Excellent understanding of business operations and analytics tools for effective analyses of data.
• Base SAS, SAS E - miner, SAS UNIX SAS, and SAS Enterprise Guide, SPSS, FICO Modeler, SQL and R.
• Specializes in delivering Modeling and Validation solutions mainly on Credit Risk (PD/EAD/LGD, CCAR/DFAST, Stress Test, Economic Capital, Regulatory Capital & Reporting), VaR, Scorecard, Collection and Data analytics modeling using Predictive Analytics - Multiple Regression, Logistic Regression, Time Series Analysis (ARIMA, VARMAX, and regARIMA), and Scorecard.
• Adaptive to the team environment and has ability of completing complex tasks independently.
• Highly adaptable to fluctuating requirements, situations and excellent communication skills.
• Expertise in delivering solutions in Text Mining, Fraud Analytics using R
• Built Dashboards using SAS reporting.

TECHNICAL SKILLS

Domain: Banking and Retail.

Technical dynamics: Base-SAS, SAS enterprise guide, SAS E-miner, SAS ETS, SAS Forecast Studio, SAS Text analytics, R project, SPSS, MS SQL Server, MS- Access, Tableau

PROFESSIONAL EXPERIENCE

Senior Lead Analyst-Data Science

Confidential, SunTrust Bank VA

Responsibilities:

• Developed the Auto loan default prediction model for consumer lending business.
• Developed Fraud detection models using R.
• Developed recovery dashboards for the default tracking.
• Developed, validated the Light Vehicle Sales Current Activity Index Model. (The model) estimates a current activity index (CAI) for new light vehicle sales (LVS) volume to provide insight into the relationship between economic indicators and LVS volume. In this model, principal component analysis summarizes the information contained in multiple economic variables into one observable statistic variable.
• Developed, validated the Macroeconomic VAR Model using SAS Autoreg, regression VARMAX procedure.
• The Light vehicle sales model for Auto loans was developed using X-12 ARIMA methodology. The Light Vehicle Sales Seasonal Factors model (the model) estimates the monthly seasonal factors for new light vehicle sales. The estimated seasonal adjustment factors (SAF) are used to translate the seasonally adjusted annual rate (SAAR) of light vehicle sales to non-seasonally adjusted (NSA) light vehicle sales for use in the financial planning and analysis monthly baseline process, the Comprehensive Capital Analysis and Review (CCAR), and the Dodd-Frank Act Stress Test (DFAST).
• The bankruptcy forecast filing model was built to produces a 31 month forecast of expected monthly bankruptcy filings for retail automotive Line of Business (LoB) using SAS Autoreg, regression ARIMA, ARIMAX.
• Developed and performed the performance evaluation of the scorecard developed for the auto loans. The objective was the performance evaluation of the scorecard developed for the auto loans. PSI, ROC, Gini coefficient was applied for the performance evaluation to check the performance of the new data.
• Developed the Collections Behavior and Roll Rate Scorecards used SAS, FICO Model Builder in the model development.
• Developed the regression model to predict the sales proceeds to forecast end-of-term auto lease’s sales proceeds (percentage of the base residual value specified at contract inception).
• Validated and lead the team in the validation of PD LGD models.
• Built text analytics models for feedback analysis using R and presented the results in the Tableau.
• Built logistic regression models using R.

Analyst

Confidential

Responsibilities:

• Used statistical technique Market basket analysis, Text mining, regression and forecasting for the retail domain clients.
• The forecasting model was developed using the ARIMA modeling methodology to predict the commodity prices.
• The probability to default model was developed for default prediction of existing customers.
• Optimized data collection procedures and generated reports on a weekly, monthly, and quarterly basis.
• Used advanced Microsoft Excel to create pivot tables, used VLOOKUP, and other Excel functions.
• Build the forecasting models.Lead the team of the modelers working on the modeling projects.
• Development and regular production of Management information system (MIS) or customer analysis report with the added responsibility of increasing efficiency and adding recommendation and meaningful insights to help decision making.
• Customer profiling based on performance and demographic data.
• Analyze campaigns and provide target population to increase response rate.
• Long term projects like Customer Attrition analysis and scorecard and internet opportunity to identify the target population for retention, spends campaign, high erosion customers using the Logistic regression and regression modeling.

Business Analyst

Confidential

Responsibilities:

• Development of SAS codes for reports, executing the statistical models and reporting the key takeaways.
• Understand clients' detailed requirements, set appropriate delivery expectations and deliver high level of services.
• Mentoring and guiding team members in their project delivery.
• Ensure required project and process documents are prepared, updated, shared, and maintained in secured folders.
• Development and regular production of Management information system (MIS) or customer analysis report with the added responsibility of increasing efficiency and adding recommendation and meaningful insights to help decision making.
• Generate insights and improve decision making and risk assessment of customers, using the recently acquired external trades’ data (SBMA) from Equifax.
• Analysis study of overall customers on the basis of delinquencies, revolving balances and balances
• Generate insights about current customer’s performance on external trades, eg how many accounts have financial/non-financial trade(s) outside the current business.
• Re-estimate internal models using SBMA data and get a measure of incremental lift (if any) provided by SBMA variables using the logistic Regression modeling Approach.
• Developed models using logistic regression approach to predict the charge off customers.