Thanks for a great Analytics and Data Summit 2019 event. We hope to see you next year Feb 25-27, 2020!
To download the presentations you must first sign up with Sched for free (easy and fast!)
  • Sessions appear in the color of their primary track and can be filtered using Products on the right
  • Use the Search bar for more flexibility
See this link for hints on how to search the schedule

Back To Schedule
Thursday, March 14 • 12:00pm - 12:30pm
Identifying Portfolio Variables using Machine Learning for developing Mortgage Prepayments predictor models in Banking and Finance

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Portfolio Variables are variables which identify the fingerprint of a dataset. It is most advisable to separate a dataset into multiple dataset based on values of portfolio variables and build separate models since the relationship between the dependent variable and independent variables differs across the spectrum of portfolio variables.

Portfolio variables are generally identified by Subject Matter Experts in the Industry. However with changing datasets and large number of categorical and numeric variables, Machine Learning can go a long way in adding to such portfolio variables to achieve higher prediction accuracy.

We in OFSAA team have built an algorithm to identify and segregate such variables using ML techniques which will be a part of presentation.

avatar for Jayesh Gokhale

Jayesh Gokhale

Snr Principal Software Developer, Oracle Financial Services Software Limited
Senior Principal Software Developer Development Lead - OFSAA Performance Analytics Suite     Worked on developing models for IPA/RPA, RCA and LLFP products. ML Techniques used in the models include Regression, Clustering, Time Series Forecasting, Neural Networks, Survival Models and Vintage M... Read More →

Thursday March 14, 2019 12:00pm - 12:30pm PDT
6-Rm 203