Industry Interface  

The center focuses on doing academic research with an industry application rigor. We design business tools using sophisticated models based on our ongoing research. These tools are actively used by lending institutions, credit bureaus and regulatory bodies whom we associate with as knowledge partners.  


Listing few of our recently designed tool:  

  1. Localized Distress identifier Dashboard 
  2. Alternate Credit Scoring for New to Credit (NTC) customers  
  3. Customer Retention tool for Banking Correspondent Aggregators  
  4. Anomaly Detection in Transaction through Banking Correspondent Aggregators 


Localized Distress identifier Dashboard:  

The tool is designed to indicate localized distress which may go unidentified. The tool uses MNREGA demand for work and night light radiance as the basis to determine whether the area is under distress or not.  

The tool has been designed based on the paper titled "A Friend Indeed: Does the Use of Digital Identity Make Welfare Programs Truly Counter Cyclical?". 


Alternate Credit Scoring for New to Credit (NTC) customers  

The tool uses state of the art machine learning algorithms to give a risk score associated with each new loan applicant. This tool uses information from the customer application, economic indicators local to the demographic factors and loan repayment partners of existing customers. 

The tool has been designed based on the paper titled " Fintech for the Poor: Financial Intermediation Without Discrimination". 

 
Customer Retention tool for Banking Correspondents 

The tool uses state of art machine leaning algorithm to predict the stickiness of a customer to the banking correspondent network after a failed transaction. The tool helps the banking correspondent agents to take corrective measures to retain customers who are likely to drop out. 

The tool has been designed based on the paper titled " Fintech for The Poor: Do Technological Failures Deter Financial Inclusion?". 

 
Anomaly Detection in Transaction through Banking Correspondents  

The tool uses cluster models along with basic statistical predictors to identify volatility in banking correspondents’ transaction pattern. The tool helps the aggregator to monitor major deviation in agents’ transaction patterns.  

The tool has been designed based on the paper titled " Fintech for The Poor: Do Technological Failures Deter Financial Inclusion?".