Bayesian Nonparametric Inference in Bank Business Models with Transient and Persistent Cost Inefficiency
Paper presents Bayesian nonparametric methodology to model dynamic bank business models, inefficiencies, and transitions, using empirical EU bank data.
What it examines
This paper proposes a new Bayesian model to classify bank business models using a nonparametric approach and a stochastic frontier framework. It separates cost inefficiency into persistent and transient parts while applying mixture modeling and dynamic clustering to track changes over time and adjust to economic shocks.
What it concludes
The study shows that banks often change business models following major shocks, affecting cost efficiency. The results help regulators and bank managers understand strategic shifts and efficiency gains. Future research could refine these models and explore broader applications in banking, financial stability, and policy-making.
Evidence objects
The study introduces a dynamic Bayesian nonparametric framework using the Logit Stick-Breaking Process, distinguishing transient and persistent cost inefficiencies and showing banks switch models during major shocks like COVID-19 recently.
key_findings bullet 1 · key_findings · validation V0
The paper presents state-of-the-art techniques including advanced MCMC methods and extensive simulation studies, although its reliance on large datasets and technical complexity may restrict immediate practical application for industry practitioners.
key_findings bullet 2 · key_findings · validation V0
Innovative terminologies such as transitory efficiency and persistent efficiency enrich economic discussions on cost management, with this methodology paving the way for future exploration of model uncertainty and banking applications.
key_findings bullet 3 · key_findings · validation V0
This paper presents a groundbreaking Bayesian nonparametric framework that dynamically classifies bank business models by isolating persistent and transient cost inefficiencies. Utilizing the Logit Stick-Breaking Process for predictor-dependent clustering, it offers a novel econometric methodology. Its innovative approach provides fresh insights, appealing greatly to researchers in computational and quantitative finance.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
- … in response to changing financial and economic conditions… methods with efficient computational routines. We apply … shocks, such as the global financial crisis, by switching …
Source row: 288 · abstract type: snippet