AI Case Study
Citi reduced cost, time and reliability issues in the delivery of the complex stress test models needed for regulatory compliance through deploying ML to support the process
Citi reduced cost, time and reliability issues in its delivery of the complex stress test models needed for regulatory compliance. Citi tackled the stress testing using model acceleration and improvement technology from Ayasdi.
"Citi selected Ayasdi to supplement its capital planning process. The process began with the leaders of the bank’s business units reviewing the macroeconomic variables stipulated by the Federal Reserve.
Ayasdi enriched these variables using several techniques (e.g., time series transforms such as lags, differences, and percent changes) and created over two thousand variables. Ayasdi applied its machine intelligence software to rapidly correlate and analyze the impact of these variables on each business unit’s monthly revenue performance over a six-year period, uncovering statistically significant variables that were highly correlated with each business unit’s performance.
A comprehensive business review was conducted to screen the identified variables prior to inclusion in the models for each business unit. Ayasdi then conducted exhaustive statistical tests (including stationarity and multicollinearity tests) to validate these models’ ability to predict revenues for the business units. The business leads then evaluated the candidate model and the challenger models, selecting those that best represented their business units. With a collection of accurate, transparent and defensible revenue forecast models, the bank was able to easily clear their most challenging regulatory hurdle.
Ayasdi claims: "The process compressed the resources required from a nine-month process requiring hundreds of employees to a three-month sprint with less than one hundred."
Legal And Compliance
According to Ayasdi:
"Citi consistently struggled to pass its annual stress test, failing two of the first three stress tests. The bank was in need of a way to rapidly create accurate, defensible models that would prove to the Federal Reserve that they could adequately forecast revenues and the capital reserve required to absorb losses under stressed economic conditions. The bank’s modeling approach left the business unit leads with little room and time to weigh in on the logic behind the choice of the variables selected. The result was the firm could not confidently defend the models that they included in the filings they presented to the Federal Reserve."