Newest papers

Which Text Matters: Handling Model Uncertainty in Text-as-Data Regressions, with an Application to Central Bank Communication

Martin Feldkircher, Bernd Prostmaier, Paul Hofmarcher, Pierre Siklos · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.19720 · cited by 0 works (OpenAlex) · open-access PDF

German

Sozialwissenschaftler konvertieren routinemäßig Text in Covariates wie Themenanteile, ideale Punkte oder Sentiment-Score und fügen sie in statistische Modelle von Interesse ein. Dieses zweistufige Design ist bequem, aber es wird selten von der Theorie geleitet: wenig sagt dem Analysten, welche textbasierten Covariates zu konstruieren oder einbeziehen. Jeder Ansatz verkörpert umstrittene Messwahl, und konkurrierende Operationalisierungen desselben Konstrukts können unterschiedliche Schlussfolgerungen unterstützen, so dass eine einzige Forschungsfrage zu einem Menü konkurrierender Regressionsmodelle führt. Welche Text konstruierten Covariates gehören zu einem Modell und in welcher Funktionalität, ist daher ein Problem der Modellunsicherheit. Wir schlagen Spike-and-Lab generalisierte Add

Machine translation (Meta M2M100); the English original follows.

Abstract

Social scientists routinely convert text into covariates like topic shares, ideal points or sentiment scores, and insert them into statistical models of interest. This two-step design is convenient, but it is rarely guided by theory: little tells the analyst which text-based covariates to construct or include. Each approach embodies contestable measurement choices, and competing operationalizations of the same construct can support different conclusions, so a single research question gives rise to a menu of competing regression models. Which text constructed covariates belong in a model, and in what functional form, is thus a problem of model uncertainty. We propose spike-and-slab generalized additive models (SSGAMs), which jointly select covariates and determine whether their effects are linear or nonlinear. We illustrate the approach with the communication of a powerful unelected institution, asking which of 45 competing text indicators from 224 monthly European Central Bank press conferences, spanning late 2002 to mid-2025, track professional inflation expectations. A sparse set is selected; widely used alternatives, and every tested nonlinear effect, carry no incremental information.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Covariate, Construct (python library), Computer science, Set (abstract data type), Econometrics, Ideal (ethics)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange