Semantic analysis of management discussion and analysis in 10K reports

Published: 6 December 2024| Version 1 | DOI: 10.17632/gm7xh8z5x8.1
Contributor:
igor semenenko

Description

Item 7 data processed from SEC Edgar annual 10K files in 1995-2022. Output includes sentiment scores and breakdown by parts-of-speech

Files

Steps to reproduce

Item 7 data was drawn from 10K reports using Edgar-crawler code from Lefteris et al (2021). I employ TextBlob library and Valence Aware Dictionary and sEntiment Reasoner (VADER), a module in the nltk.sentiment Python library, to analyse item 7 – Management Discussion and Analysis – in annual reports filed by the US publicly traded companies

Institutions

Acadia University

Categories

Natural Language Processing, Natural Language Semantics

Licence