Audit Analytics was honored to participate at another successful Annual Meeting of the American Accounting Association. It was great to see so many friends and colleagues and talk about the latest research in the accounting field.
Lots of professors this year were asking about using Audit Analytics in the classroom. We know many of our subscribers have developed classwork that incorporates Audit Analytics data and research. This is something we are excited about and would love to hear any feedback from users about classroom application. If you have any suggestions or questions about using Audit Analytics as a teaching aid, please feel free to contact us.
Another discussion that we heard frequently was that more and more universities are looking to introduce their students to “big data” analytics. Of course, performing such analytics requires having access to data, so many professors were interested in using Audit Analytics, especially in the business school context.
As usual, lots of interesting papers were presented during the session, many of which use Audit Analytics as a source for their research. The speech by PCAOB board member Jeanette Franzel, which cites Audit Analytics throughout, has been picked up by major accounting news sources.
Other topics discussed at the conference varied from Challenges in Auditing Accounting Estimates to the relationship between the use of company-specific XBRL extensions and financial reporting weaknesses. Featuring a representative from both KPMG and Deloitte and two from the PCAOB, the session on auditing accounting estimates offered some interesting insight from the practitioners and regulators.
As one of the panelists pointed out, virtually every number on the face financials involves estimates to one degree or another. Even cash, say in the case of a company with many foreign operations and different currencies, would require certain arbitrary decisions and estimates. With the recent focus being placed on enhanced disclosures of estimates, there is, perhaps, an expectation among users of the financial statements of a level of precision that may simply not exist.
One fascinating area of emerging research focuses on the use of linguistic and textual analysis to detect accounting irregularities. A few years ago, “Detecting Deceptive Discussions in Conference Calls” examined whether extreme positive emotion has been associated with deceptive management.
This year, during one of the presentations we were able to attend, a paper was presented that used a different approach to textual analysis. The paper, called “What Are You Saying? Using Topic to Detect Financial Misreporting”, was first presented at the mid-year FARS Section. Instead of using specific words known to be associated with deceptive management, the authors used the Latent Dirichlet Allocation methodology (LDA) to identify “topics” discussed in financial statements. According to the authors, “topics” are strongly correlated with financial misreporting (as measured by AAER’s and restatements associated with accounting irregularities).
We are always eager to hear about new research, but unfortunately this year we were only able to sit in on a couple of presentations. Check our Academic Research page for a selection of papers that cite Audit Analytics data.