A0535
Title: A topic model for 10-K management disclosures
Authors: Minh Tri Phan - University of St. Gallen (Switzerland) [presenting]
Matthias Fengler - University of Sankt Gallen (Switzerland)
Abstract: The topics discussed in the management's disclosures and analysis (MD\&A) section of 10-K filings from 1994:01 to 2018:12 are investigated. In the modelling approach, the MD\&A topics are elicited by clustering words around a set of anchor words that broadly define a potential topic. Given the topics, two hidden loading series are extracted from the MD\&As - a measure of topic prevalence and a measure of topic sentiment. The results are three-fold. First, the topics found are intelligible and distinctive but potentially multi-modal, which may explain why classical topic models applied to 10-K filings often lack interpretability. Second, topic prevalence and sentiment tend to follow trends which by and large can be rationalized historically. Third, sentiment affects topics heterogeneously, i.e., in topic-specific ways. Adding to the extant document-level techniques, the potential benefits are demonstrated for using a nuanced topic-level approach to analyze the MD\&A.
