Samuel Borms | Sentometrics Research

The R package sentometrics to compute, aggregate, and predict with textual sentiment

Sun, 01 Aug 2021 00:00:00 +0000

Textual sentiment can be computed in many ways, and pooled into a time series in many more. The package treats that multiplicity as the problem to be solved rather than a nuisance: it computes sentiment, generates many candidate aggregation schemes, and selects among them within the prediction step.

See the package site.

The economic policy uncertainty index for Flanders, Wallonia and Belgium

Tue, 01 Dec 2020 00:00:00 +0000

The index is derived daily from major Belgian newspapers in French and Dutch since 2001, using triplets of economy / policy / uncertainty keywords defined by a combination of expert judgment and NLP. A 2021 revision made the index more local to Belgium and better aligned across the two languages.

Download the index.

Sentiment and Econometrics: Toward a Unified Framework of Textual Sentiment Analysis for Economic and Financial Applications

Tue, 01 Sep 2020 00:00:00 +0000

Supervised by David Ardia and Kris Boudt. Samuel is the author and maintainer of the sentometrics R package.

Econometrics meets sentiment: An overview of methodology and applications

Mon, 01 Jun 2020 00:00:00 +0000

The reference entry point to sentometrics. It walks through the full chain of an econometric analysis of textual sentiment — corpus construction, sentiment computation, aggregation into time series, and use in a model — and organises the applications across economics, finance and politics.

Start here, then read the software vignette.