Mitul Bahl Marwah

Understanding Clients business and requirements and interpreting briefs by thinking analytical solutions to answer business questions. Work closely with teams for delivery of projects to the clients ensuring quality deliverables. Design, develop and deploy data driven predictive models to solve business problems using appropriate machine learning and statistical modelling techniques. Some of the projects worked on: Developed model to identify and prioritized certain specific type of vehicles for sales which are pending and taking high turnaround time with the salvage vendor.

Key objective.

• To Pre-process data of more than 1 Million Records and hundreds of Features.
• To build and Train Machine Learning Model using Logistics Regression, Random Forest Classification
• To find accuracy of Model Prediction using Classification Reports, Confusion Matrix, AUC Score.
• Generate actionable insights from data and creating presentations to make recommendations for improvement. Complete model was build in using python with production ready code from scratch.

Active delinquent loans in a particular month can move to any three possible stages in next 12 months: Claim, natural cure and workout. The objective of the delinquent account status prediction model is to identify a loan moving into above mention three different segments in next 12 months.

Key objective

• Basic objective was to simultaneously predict different probability of loan moving to different segment in next 12 months.
• Define relationship between categorical dependent variable and continuous/categorical independent variables. Multi nominal logistic regression technique was used to identify loans moving to particular segment with definite prob.Value.
• Saved $ value by implementing cost-saving initiatives that addressed long-standing problems.

US Mortgage Insurance organization seeks to establish an application forecast. Applications are steered by movement in macro- economic factor combine with internal policy decision.

Key objectives

• Forecast monthly application volume for better resource planning.
• Build a hybrid forecasting model, combining Time Series and ARIMA techniques.
• Forecasts of MI applications for next 7 months.
• Since macro-economic factors play big role in steering this numbers, ARIMA technique use to predict monthly numbers, taking in account of effect of existing home sales, fixed rate interest, consumer confidence index.
• Weighted Average of TS component and ARIMA is applied to predict final monthly numbers, where weights are decided by proportion of individual MAPE of these two-mention techniques.

To provide lenders with anticipated turn time of 4H and 9H application when they submit the data. Business was looking for growth rate driven by customer awareness and experience

Key Objective:

• Data had Outlier points and had to build predictive model to handle outlier's points. Used robust regression technique to build model
• Created hourly and weekday level matrix to check variance. Also, performed analysis and identify at which level we are getting higher variance.

US Bank Deposits Churn Prediction A leading US banking and financial institution had a large number of long-term deposits up for maturity. Business problem was whether the CD's will attrite after maturity or will get renewed.

Key objectives

• Predicting product based retail customers who have high likelihood to attrite on the product after the maturity.
• Logistic Regression model was used to predict likelihood of the customers who have high probability of churning after maturity in advance which helped the business to implement prevent strategies to retain these customers.

US Mortgage Insurance organization seeks to establish an application forecast.

Key Objective:

• Forecasting number of applications and reduce error rate which was 15% with current ARIMA model.
• Time series Forecasting for number of applications was performed using LSTM (Long Short-Term Memory networks) models wherein error rate reduced from 15% to under 10%.

Dashboarding and Reporting Projects for leading insurance company for tracking Accounts Receivables and operations metrics

Key Objective:

• Presenting data insights using different data visualization tools as per client requirements such as Power BI and DAX, Tableau, and Qliksense Reporting.
• Collecting and aggregating data from multiple internal and external sources for analysis to drive insights into business performance. Expertise in using advance level calculations on data set.
• Mentoring and training junior team members with tools and techniques.
• Hire suitable team members keeping strategic needs in mind.

The job responsibilities involve developing credit and marketing scoring solutions. Working with client’s analytics and Management team to provide them with insights drawn from the data through reports and predictive models. Detecting and resolving technical issues related to SAS processes.Providing Training and Support to junior team members regarding SAS processes.Assisting Team and client with various data analysis and development of models.Analyze, test, document, and maintain SAS programs and macros to generate SAS datasets, spreadsheets, data listing, tables and reports.

Some of the Key projects are :

Churn prediction : Leading telecom provider was facing a customer churn problem. Key objective:

• Analyze customer-level data of the firm, build predictive models to identify customers at high risk of churn and identify the main indicators of churn.
• Applied Logistic Regression Model to predict subscribers who are disconnecting the services and 90 day prediction in advance so that they could be retained by offering some suitable plans

Plan Design for Customer Satisfaction : Leading telecom provider was facing plan switch problem resulting in lower customer satisfaction. Key objective:

• Predict Subscribers who have high likelihood to plan switch in advance.
• Applied Logistic Regression Model to predict Subscribers who have high probability of plans switch in advance so that business can offer them the plan according to their needs and increase customer satisfaction.
• Performed Text mining in R to classify the customer feedback or comments into positive, negative and Neutral and used this classification to enhance the Model Performance.

Cross Sell Opportunity for leading telecom provider to increase the Average revenue per user.

• Applied Association analysis techniques to discover patterns in data for frequently used plans and identifying new cross sell opportunities for retaining customers.
• Worked on Customer Segmentation and Profiling based on high probability to churn and high Revenue Share to identify the target Population for marketing campaigns.

• The job involved working in close liaison with the client’s Data Intelligence team to provide descriptive analytics to solve complex business problems through data extraction and analysis.
• Following are the Key Functional Areas:-.
• Developed a predictive model through logistic regression to identify the default customers using the scores.
• Performed data mining analysis on money transactions & Behavior-based Modeling.Created a high-quality statistical-grade data set, optimized for Modeling & Analytics, emphasizing data validation, data mining, data cleansing & data reduction.
• Providing Project and SAS Training to new joinees in the team.
• Analyzed and assessed campaign effectiveness using SQL and SAS techniques.
• Subsequent extrapolation of relevant information/data from huge and complex databases.
• Developing a retention dashboard solution to have greater visibility of customers across various business segments.
• Integrating Post-Campaign data from different sources and analysing campaign effectiveness by response rate metrics and offers opted in.
• Supporting Customer Experience Metrics based on Inbound and Outbound Calls.
• Automated procedures in statistical analysis, data reporting, statistical forecasting & pattern recognition to minimize costs and maximize profits using SAS, SAS Macros, crontab.
• Segmented Banks Customer Base using their transactional data and RFM analysis to identify most valuable customers in each segment and helped marketing group to be proactive in identifying the reward/retention and growth opportunities.
• Performed Market Basket Analysis on the transaction level data to recommend the next best product for the customers depending on which cluster they fall into using clustering technique.
• Performed Call Reason analysis which helped client to work on the complaint reasons which in turn helped the client in reduction of contact centre burden and increased client satisfaction.Supplied multiple cross lines of business groups with adhoc analytical reports, presentations, data, and insights tailored to improving the customer experience through operational enhancements.

• The job involved preparing reports and generating insights, reviewing, analysing and validating data to ensure consistency, integrity and accuracy.
• Translating business needs into data management objectives.
• Providing information and analysis in order to improve marketing campaigns and other company processes.
• Highlights are as follows:-.
• Work involved data cleaning to reporting and deep dive analysis using SAS, SQL etc.Analysed campaign responses, worked with data to develop customer profiles, analyse research results.
• Developed complex queries and reports in SQL and Access.
• Generation of SQL scripts to manage data.
• Generated reports with insights through Excel and Oracle.
• Provided analytics support to marketing and finance teams to help improve overall business performance.
• Developed innovative ways to search meaningful patterns, trends & relationships by sifting through large amounts of data.
• Worked across business units & functions for acquisition & analysis of data.
• Formulation of proposals, preparation of reports and recommendations on policy and trends.
• Handled a team of 3 members and guided them in preparation of reports.
• Worked on Various reports involving Customer churn reports by demographics, usage etc.
• Customer Profiling based on demographics data, Analysis on Revenue data and Call details data and Identification of customer segmentation using revenues and CLV data.

• The job involved provision of support to Data Retrieval Applications to ensure data extraction in a timely and precise manner.
• Design, program, document, and maintain SQL queries necessary to prepare data, based on complex project requirements and variations in data sources/formatting.
• Provide weekly and monthly analysis to include identification of positive and negative trends for the project.
• Analyze data fields, design and produce reports in Excel, make recommendations to improve the project performance.
• Resolve data discrepancies and ensure data accuracy.
• Use SQL to create and conduct Quality-control checks on data at all stages.
• Create and Execute efficient database queries to provide results to business units.
• Extracting data from large database by SQL Queries in Toad and validating and verifying the data for the project.

• The aim was to develop, deploy, and provide data for solving business problems.
• Preparing adhoc reports using SQL to support sales and marketing team with new prospective markets for the sales of product in terms of quantity and revenue.
• Analysing requirements for business reporting (data extraction and manipulation) as well as report design, development and delivery.
• Monitor Competitors, and effectively incorporate Competitive Intelligence into business.
• Maintain database and find Opportunities and Prospects for the industry.
• Collect, interpret, and provide industry, company, competitor, and market segment profiles, trends, and overviews.
• To perform research and analysis of various Industries by doing wide search using paid databases such as Bloomberg, D&B etc and also the Internet.
• Conducting secondary research on energy and power data to provide accurate and meaningful information for market planning, sales, pricing, costing and forecasting for retail markets and various business issues.

• The role involved extraction of data from existing database by performing queries & procedures in SQL and PL/SQL.
• Developed and maintained procedures to extract and validate data from SQL Server databases and generate reports in Excel according to clients Requirements.
• Uploading of data collected through various sources from text or excel files to SQL Server database.
• Data gathering, maintenance, extraction, cleaning, reporting and Business Development related Research.