Percentage of word occurrences in MD&A sections for fraudulent vs. non-fraudulent companies. Note: The y-axis represents the average yearly percentage of each word category occurring in the MD&A sections. The word categories (Positive, Negative, Uncertainty, and Litigious) are based on the classifications by Loughran and McDonald (2023).

 
 
  Part of: Ketelaar F, Mićković A (2025) Artificial Intelligence in fraud detection: textual analysis of 10-K filings. Maandblad voor Accountancy en Bedrijfseconomie 99(2): 61-71. https://doi.org/10.5117/mab.99.132881