Abstract
Investigators publish new report on Ma chine Learning. According to news originating from Montreal, Canada, by NewsRx e ditors, the research stated, "This paper assesses the utility of machine learnin g (ML) techniques combined with comprehensive macroeconomic and microeconomic da tasets in enhancing risk analysis during stress tests. The analysis unfolds in t wo stages." Financial supporters for this research include Fonds de recherche sur la societe et la culture (Quebec), Canadian chair in Macroeconomy and Forecasting (UQAM). Our news journalists obtained a quote from the research from the University of Q uebec Montreal, "I initially benchmark ML's efficacy in forecasting two pivotal banking variables, net charge-off (NCO) and pre-provision net revenue (PPNR), ag ainst traditional linear models. Results underscore the superiority of Random Fo rest and Adaptive Lasso models in this context. Subsequently, I use these models to project PPNR and NCO for selected bank holding companies under adverse stres s scenarios. This exercise feeds into the Tier 1 common equity capital (T1CR) de nsities simulation. T1CR is the equity capital ratio corrected by some regulator y adjustments to risk-weighted assets. Crucially, findings reveal a pronounced l eft skew in the T1CR distribution for globally systemically important banks vis- & agrave;-vis linear models."
NETL
NSTL