Search Results
Working Paper
Explaining Machine Learning by Bootstrapping Partial Marginal Effects and Shapley Values
Machine learning and artificial intelligence are often described as “black boxes.” Traditional linear regression is interpreted through its marginal relationships as captured by regression coefficients. We show that the same marginal relationship can be described rigorously for any machine learning model by calculating the slope of the partial dependence functions, which we call the partial marginal effect (PME). We prove that the PME of OLS is analytically equivalent to the OLS regression coefficient. Bootstrapping provides standard errors and confidence intervals around the point ...
Working Paper
Impact of the Volcker Rule on the Trading Revenue of Largest U.S. Trading Firms During the COVID-19 Crisis Period
Using a novel data collection, we examine the impact of the Volcker Rule on trading revenue of the 21 largest U.S. trading firms during the 100 day stress period centered on the COVID-19 financial crisis. We find that despite the market volatility, trading profits were consistent with volume-driven fees, commissions, and widening of the bid-ask spread. This work adds to the growing body of evidence that a consequence of the Volcker Rule on firm revenue associated with trading is increased financial stability and decreased risk exposure to market shocks.
Discussion Paper
Using Service Provider Connections to Model Operational Payment Networks
This paper uses data on bank connections with service providers to construct a representation of an operational network used to facilitate the sending of Fedwire transactions. Our data contains 227 connections between 215 banks (mostly community banks, but also some large banks) and four unique payment products used by the firms to send and receive Fedwire transactions. By constructing such an operational network between banks and payment providers, we can perform multiple analyses that are useful in operational resilience considerations. First, we use the mean daily Fedwire volume for each ...
Working Paper
Explaining Machine Learning by Bootstrapping Partial Dependence Functions and Shapley Values
Machine learning and artificial intelligence methods are often referred to as “black boxes” when compared with traditional regression-based approaches. However, both traditional and machine learning methods are concerned with modeling the joint distribution between endogenous (target) and exogenous (input) variables. Where linear models describe the fitted relationship between the target and input variables via the slope of that relationship (coefficient estimates), the same fitted relationship can be described rigorously for any machine learning model by first-differencing the partial ...