SentoResearch

Research material & open-source software for the community

Browse the indices Read the survey paper

In a nutshell

Within the growing landscape at the frontier of text mining, sentiment analysis, and econometrics, the field sentometrics has emerged. Researchers in sentometrics study how the qualitative sentiment embedded in textual data becomes a quantitative variable — and how that variable behaves once it enters an econometric analysis.

This hub is where our team publishes what it builds. Not a directory of the field: search engines do that better, and a list that falls behind misleads. What you will find here is what we produce and keep alive — the indices, the papers behind them, the packages that compute them, the students who work on them, and everything we do to make all of it reusable.

New here? Read the survey paper, then the R package sentometrics. If you want to work with us, start here.

Indices

MCCC

Daily U.S. Media Climate Change Concerns Index, with topical and thematic subindices. January 2003 to June 2025.

EPU Belgium

Daily economic policy uncertainty for Flanders, Wallonia and Belgium, built separately from French- and Dutch-language sources.

EPU Quebec

Monthly economic policy uncertainty for Quebec, built from French-Canadian newspaper archives. A century of data.

U.S. Topical Economic Sentiment

Daily sentiment indices for topics relevant to the U.S. economy, computed with several lexicon-based methods.

Packages

sentometrics

The full workflow in one package: compute sentiment, generate many candidate aggregation schemes, and select among them within the prediction step.

sentopics

Estimating and analysing sentiment and topic models — for when the index needs to be topical rather than aggregate.

sentometrics.app

A Shiny interface to sentometrics, for exploring aggregation choices without writing code.

GWP

Calibrating a sentiment lexicon against a target with the Generalized Word Power methodology, instead of assuming word weights.

Papers

These are the entry points. See all our papers — filterable by type and year.
David Ardia, Keven Bluteau (2026). Optimal text-based time-series indices. International Journal of Forecasting, 42(1), 44–60.

PDF Code DOI

David Ardia, Keven Bluteau, Kris Boudt, Koen Inghelbrecht (2023). Climate change concerns and the performance of green versus brown stocks. Management Science, 69(12), 7607–7632.

PDF Dataset DOI

David Ardia, Keven Bluteau, Samuel Borms, Kris Boudt (2021). The R package sentometrics to compute, aggregate, and predict with textual sentiment. Journal of Statistical Software, 99(2), 1–40.

PDF Code DOI

Andres Algaba, David Ardia, Keven Bluteau, Samuel Borms, Kris Boudt (2020). Econometrics meets sentiment: An overview of methodology and applications. Journal of Economic Surveys, 34(3), 512–547.

PDF DOI

David Ardia, Keven Bluteau, Kris Boudt (2019). Questioning the news about economic growth: Sparse forecasting using thousands of news-based sentiment values. International Journal of Forecasting, 35(4), 1370–1386.

PDF Dataset DOI

Team

Sentometrics is run by a small team across HEC Montréal, Université de Sherbrooke and Vrije Universiteit Brussel.

Researchers

Avatar

David Ardia

Professor of Quantitative Methods

Avatar

Keven Bluteau

Associate Professor of Finance

Avatar

Kris Boudt

Professor of Finance and Econometrics

Students & alumni

Sentometrics is also a training programme. Since 2016 the field has produced doctoral theses, master’s theses and open-source packages across five universities in three countries. This is the lineage.

Doctoral researchers and postdocs

Master’s theses

Student co-authors

Alaa Kassem (EPU Quebec, Economics Letters 2021) · Mohammad Abbas Meghani (Thirty years of academic finance, Journal of Economic Surveys 2024) · Thien Duy Tran and Thomas Lortie-Cloutier (green and brown stocks, Finance Research Letters 2022 and 2023).

Beyond these, the sentiment and NLP strand of more than sixty supervised master’s projects — many with industry partners — has been built on the indices and software published here.

Every thesis above links to its entry on this site, which carries the full text.

Want to work on sentometrics? See how to get started.

Knowledge transfer

Everything here is built to be reused. The indices are free for academic research, the packages are on CRAN and GitHub, and the methodology is documented in open-access papers.

  • How to cite — ready-to-copy BibTeX for every index, package and method paper.
  • Working on sentometrics — where to start, which data to build on, and the topics we think are worth a thesis.
  • Using one of our indices? Tell us — we keep a public record of the research they support.

Talks and outreach

In the press

Recognition

Research professorship in Sentometrics (HEC Montréal, 2020) · National Bank of Belgium research award 2020 · Best paper 2018–2019, International Journal of Forecasting · Open data quality award, Canadian Open Data Society, 2024 · Research impact award, HEC Montréal, 2025