Search Results

You are looking at 1 - 10 of 180 items for :

  • "bears risk" x
Clear All
Majid Bazarbash

Gradient Boosting Trees Support Vector Machines (SVMs) Neural Networks C. How Does Machine Learning Differ from Econometrics? IV. Strengths and Weaknesses of ML-Based Lending for Financial Inclusion A. Strengths of ML-Based Lending ML can make assessing credit risk of small borrowers feasible and economical ML can harden soft information ML can better capture nonlinearities ML could mitigate information asymmetry B. Weaknesses of ML-Based Lending ML-based lending bears risks of financial exclusion ML-based credit rating could cause

Majid Bazarbash

risk management practices and could result in higher profitability of lenders and potentially superior financial stability. B. Weaknesses of ML-Based Lending ML-based lending bears risks of financial exclusion ML models are trained using available data that may not necessarily be representative of all classes of borrowers that the creditor considers lending to—a situation that violates the principle of generalizability. As a result, the lack of sufficient relevant data for some classes would impose restrictions on the conclusions by ML analysis and