Newest papers

From Word Counts to Context: Topic Models for Asset Pricing

Kevin Foley, Jonathan Hartadi, Shivesh Prakash, Swapnil Vatsal · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2609.34169 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Berita browsing dapat mengungkapkan risiko sistematis, tetapi apakah konteksnya meningkatkan konstruksi faktor risiko sistematis masih belum jelas. Kami berusaha untuk menguji apakah menggunakan transformator kalimat mewakili perbaikan atas teknik seperti Latent Dirichlet Allocation (LDA) dalam koherensi daftar istilah topik yang dihasilkan dari data teks yang tidak terstruktur. Untuk menguji ini, koleksi data teks yang tidak terstruktur yang sama yang terdiri dari 394.661 artikel dan saluran pembuatan portofolio keuangan yang sama di bawah ini diterapkan dengan lapisan teks yang berbeda, termasuk panjang teks artikel setiap model yang digunakan dan bagaimana istilah topik dinilai: kami membandingkan LDA terhadap transformator kalimat beku dengan k-me clanstering. Kami menemukan bahwa cabang transformator kalimat memiliki skor yang lebih tinggi baik dalam hal koherensi (diukur oleh NPMI) serta kinerja keuangan (diukur oleh Sharpe), meskipun

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

Abstract

News may reveal systematic risk, but whether its context enhances the construction of systematic risk factors is still unclear. We seek to test whether utilizing a sentence transformer represents an improvement over techniques such as Latent Dirichlet Allocation (LDA) in the coherence of topic term lists generated from unstructured text data. To test this, the same collection of unstructured text data comprising of 394,661 articles and the same downstream financial portfolio construction pipeline were applied with the text layer differing, including the length of article text each model used and how topic terms were ranked: we benchmark LDA against a frozen sentence transformer with k-means clustering. We find that the sentence transformer branch had higher observed scores both in terms of coherence (measured by NPMI) as well as financial performance (measured by Sharpe), although the available tests do not establish outperformance. Further exploratory specifications such as utilizing spherical clustering and multi-horizon exposures had an observed excess-return Sharpe of 1.03 for the combined model. We believe that there is some promise in applying context-aware techniques on unstructured news text, but stricter tests using only information available at each date and broader datasets may be required to enhance the confidence in the observed performance.

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

Topics: Latent Dirichlet allocation, Topic model, Computer science, Portfolio, Sentence, Transformer

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