David Ardia

AI, machine learning, and NLP for finance

About

I'm a quant at heart. I build methods that turn messy data — prices, news, filings, anything written down — into signals for risk, portfolios, and the economy. Financial econometrics is my home base; artificial intelligence, machine learning, and natural language processing are how I get there. Above all, I like working with people — colleagues, co-authors, and students. That part matters to me as much as the results.

I've been shipping open-source code for twenty years. With my team I maintain a dozen R packages on CRAN — GARCH and regime-switching models, risk-based portfolios, textual sentiment, global optimization — because a method nobody can run is a method that doesn't exist. I'm also active in Sentometrics Research, bridging text mining, sentiment analysis, and econometrics, and FAME, a joint Paris Dauphine–PSL and HEC Montréal initiative on generative AI and large language models in financial markets.

If you like clicking links, here is the institutional trail: I am an elected member of the ISI, a member of CIRANO, CIREQ, CRM, Fin-ML, GERAD, OBVIA, the Penner Institute, and Quantact, and an instructor at DataCamp.

Selected publications

Book

Cover of Financial Risk Management with Bayesian Estimation of GARCH Models

Financial Risk Management with Bayesian Estimation of GARCH Models

Theory and Applications · Lecture Notes in Economics and Mathematical Systems 612, Springer, 2008

Bayesian estimation of GARCH models by MCMC, and what it buys you for risk measurement — the methods behind my bayesGARCH package.