Michele Fabi Receives Best Paper Award at the 2023 International Fintech Research Conference

Michele Fabi, a postdoctoral fellow at CREST-École polytechnique has recently received the Award for the best paper presented at the International Fintech Research Conference, held at Parthenope University of Naples on 2-3 November 2023.

Michele Fabi

Michele Fabi is currently a post-doctoral researcher at CREST-Ecole polytechnique, serving as a member of the Blockchain and Platform Chair—a research initiative dedicated to blockchains and associated technologies.

Broadly speaking, Michele’s research fields are Financial Economics and Microeconomics, with a focus on topics such as fintech, platforms, decentralized and startup finance, blockchain, and the digitalization of money and markets. He is actively involved into a promoting research on these subjects by taking part in the organization of academic activities such as the BlockSem seminar and the Blockchain@X-OMI Workshop on Blockchain and Decentralized Finance.

Finally, Michele obtained his PhD in 2021 from the doctoral program IDEA-UAB (Universitat Autonoma de Barcelona), working with Matthew Ellman as his advisor.

In 2023-2024, Michele is one of CREST’s International Job Market candidate.

2nd International Fintech Research Conference

The International Fintech Research Conference is set to be a vibrant hub for researchers, encouraging cross-disciplinary discussions and collaborations in the expansive field of Fintech. Covering diverse areas, the conference welcomes contributions in theoretical analysis, machine learning applications, cryptocurrencies, cybersecurity, neural networks, smart contracts, and more.

With a focus on cutting-edge topics such as blockchain technologies, big data analysis, and behavioral finance, the event promises to be a dynamic platform for researchers to share insights, engage in discussions, and shape the future of finance and technology.

During this conference, the Organizing Committee establishes a prize for the best paper presented which was awarded, this year, to Michele Fabi for his paper “Blockchain Design with Transmission Delays”.

More information on the Conference: https://www.disaq.uniparthenope.it/fintechlab/international-fintech-research-conference/

Blockchain Design with Transmission Delays

The article investigates the economics of blockchain design. Unlike previous studies, this research introduces a strategic variable for miners: the ability to choose the size of transaction blocks. This departure from the assumption of maximum block capacity provides insights applicable to modern blockchains like Ethereum.

The study makes two key contributions. Firstly, it establishes testable predictions on block size and justifies transaction fees from the perspective of consensus layer incentives. This differs from earlier approaches that focused on application layer incentives. Secondly, the research addresses the limitations of a partial-equilibrium approach found in previous market microstructure papers. By endogenizing token inflation, the model can provide insights into the optimal balance between transaction fees and token inflation.

The paper builds upon prior works in the economics literature on blockchain and cryptocurrencies, pioneered by Chiu and Koeppl (2019) and Pagnotta (2022) among others, incorporating elements of the new monetarist matching models like Choi and Rocheteau (2020). It is also deeply inspired by the computer science literature on blockchain consensus and distributed algorithms (e.g. Pass and Shi, 2017; Ren, 2019), which lays out the methodological and conceptual foundations for the study of miner incentives.

In conclusion, the article sheds light on the economic challenges of blockchain designs, providing valuable insights for both researchers and practitioners in the blockchain space.

To read the full paper: https://drive.google.com/file/d/1VWGl65Mbg5iCSVNWnW3IMZZ1IL4oJuEU/view

From Lab Desks to Bookshelves: ‘Machine Learning pour l’Économétrie’ Authored by Lab Alumni

We are pleased to announce the publication by Editions Economica (ENSAE Collection, Advanced Economics and Statistics) of Machine Learning for Econometrics by Christophe Gaillac and Jeremy L’hour.

Two PhD alumni from CREST

Christophe Gaillac is a Postdoctoral Prize Research Fellow in Economics at Nuffield College, University of Oxford and was previously a PhD student at CREST. His research focuses on Econometrics, Statistics, and Machine Learning. Christophe is also an alumni from École polytechnique and ENSAE Paris where he was, later, the coordinator of statistical and mathematical teachings from 2017 to 2019 at ENSAE Paris.

Jeremy L’hour works on econometric methods and applications of machine learning to causal inference. He currently works as a quantitative researcher in the statistical arbitrage alpha team at Capital Fund Management (CFM) and he holds a PhD in Economics from Université Paris-Saclay prepared at CREST-ENSAE. Jeremy is an alumni of ENSAE Paris and he was the coordinator for econometrics teachings between 2016 and 2019.

Machine Learning For Econometrics

Machine Learning pour l’économétrie” is a work intended for economists who wish to grasp modern machine learning techniques – from their predictive performance to the revolutionary processing of unstructured data – in order to establish causality relationships from the data.

It addresses the automatic selection of variables in various high-dimensional contexts, the estimation of treatment effects heterogeneity, natural language processing (NLP) techniques, as well as synthetic control and macroeconomic forecasting.

The foundations of machine learning methods are introduced to provide both a thorough theoretical treatment of how they can be used in econometrics and numerous economic applications. Each chapter contains a series of empirical examples, programs, and exercises to facilitate the adoption and implementation of techniques by the reader.

This book is aimed at master’s or graduate students, researchers, and practitioners eager to understand and enhance their knowledge of machine learning to apply it in a context traditionally reserved for econometrics.