« Because it's Friday: The speaking player piano | Main | A Review of the R Graphics Cookbook »

February 11, 2013

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a010534b1db25970b017d40f7ba97970c

Listed below are links to weblogs that reference Did an Excel error bring down the London Whale?:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

B.D. McCullough of Drexel specializes in the reliability of statistical software. He first challenged Excel's math in 1999, and has published a total of ten articles on the subject in professional journals such as Computational Statistics and Data Analysis. There is a list of publications on his home page at http://www.pages.drexel.edu/~bdm25/

Excel is not the only software with questionable math. When IBM acquired SPSS, they tested it with the NIST datasets; one-way ANOVA, GLM and MEANS all failed. IBM, to its credit, owned up to the issue. http://www-01.ibm.com/support/docview.wss?uid=swg21485458

I used to work at one of the biggest banks a decade ago, in a backoffice team doing risk management. All we used was Excel!

Whether developed with the Excel tool or some other application, Value-at-Risk is not a coherent measure, and not a sound basis for assessing the risk of an investment.

If Excel is indeed an unsound estimation tool, then the "Whale" was victim of a two-fer: an unsound risk measure calculated using an unreliable estimating tool.

Great post,

We're seeing more and more in the news about companies losing big bucks due to poor Excel skills. It's a trend that I see continuing until proper investment is made into simple training courses.

Although JP Morgans Excel docs would have been more complicated that most, there were still a number of techniques that they could have used in order to avoid these glaring mistakes. I've written about them in my most recent post for Microsoft Training dot net.

http://www.microsofttraining.net/b/exceltraining/2013/04/get-back-control-of-your-excel-spreadsheets/

A great shame. Excel is good for some things, but it does seem appropriate in some cases to teach people in the world of finance more software engineering skills, even if only to get them to write excel macros that do suitable validation....

The comments to this entry are closed.


R for the Enterprise

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid

Search Revolutions Blog