presentations
presentations
Dr. Richard M. Crowley — Tue 05 April 2022
Misreporting Detection: Past to Future

Misreporting Detection: Past to Future

This talk is part of the Hong Kong RGC IIDS lecture series on Accounting Data Analytics hosted by Hong Kong Shue Yan University.

The lecture covers:

  1. What misreporting is, and how it is measured in the US.
  2. Econometrics for misreporting detection,
  3. Older approaches: Financial ratios, F-Score, and textual style characteristics.
  4. Newer methods: Brown, Crowley, and Elliott (2020)
  5. Extension using nonparametric ML (XGBoost).

Slides are available in an interactive web-based format (reveal.js) and in PDF.

revealjs pdf

Code to replicate the analysis in R is available below (note: measure construction isn’t included). Visualizations with their code are included in an Rmarkdown HTML notebook.

rfile rnotebook

More details are available in my paper with W. Brooke Elliott and Nerissa Brown, published in the Journal of Accounting Research. A copy is available publicly on SSRN:

pdfssrn