Take the 2-minute tour ×
Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. It's 100% free, no registration required.

I am trying to develop a short equity strategy that dedicates some, currently arbitrary, percentage of short capital to potential frauds.

I am aware of Benford's Law, but this is not my specialty.

What quantitative methods are available to potentially detect fraud?

share|improve this question
Not an answer to the question, but you may be interested in extremely accurate models in the literature of default probability. But more onto your question, this is a supervised learning problem. You need data on frauds and non frauds, and all features you can possibly get your hands on, then train your model and test it using cross-validation or something of the sort. Quite a straightforward problem actually. The difficulty is getting relevant features together. This could come in the form of secondary market process characteristics, accounting variables, dummy variables (industry), etc. –  user2763361 Dec 29 '13 at 10:46
To add to the above, most likely you want a nonlinear model such as random forests since it will be interactions between these variables that are likely to indicate fraud. –  user2763361 Dec 29 '13 at 10:49
@user2763361 Thank you! I've already started down the default non-default road, but thank you very much for "random forests". That's new on me. –  quantycuenta Dec 30 '13 at 22:33
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.