September 09, 2004

IEM Analysis Spit'n'Polish Dept

Posted by Kieran

As a spin-off from Daniel’s discussion of whether the DEM04 contract is overvalued on the Iowa Electronic Markets, here’s a version of the trend surface he calculated that shows differences between the Black-Scholes valuation and the observed market price over time (you can look at it in smaller PNG format or better-quality PDF). I created it using R, the free1 statistics package because I didn’t like Excel’s default effort and I hadn’t had a reason to use R’s wireframe() function before. It’s still not up to the standards of the Bill Clevelands or Ed Tuftes of this world, but it was the best I could manage on short notice. Thanks to Daniel for sending me the data, and remember that whereas I am happy to field questions about graph colors and chart widgets, technical queries about option valuation, Black-Scholes volatility fluctuations and arbitrage should still be directed to him.

1 As in “free to make your own mistakes.”

Posted on September 9, 2004 03:31 AM UTC
Comments

” It’s still not up to the standards of the Bill Clevelands or Ed Tuftes of this world…”

So little is. I just read Tufte’s “The Cogitive Style of Powerpoint.” Wonderful. I wish everyone who presents using PP outpoint output had to read it first.

Posted by Jonathan Goldberg · September 9, 2004 01:27 PM

Well yes, I must agree, that is a bloody impressive plot.

Posted by dsquared · September 9, 2004 01:32 PM

Right, inspired by that, I have now downloaded R. Any chance of posting the R commands which created the plot?

Posted by dsquared · September 9, 2004 06:21 PM

Sure, here ya go:

The code is in surface.r and uses this CSV data file.

Posted by Kieran Healy · September 9, 2004 10:07 PM

god that’s impressive. this looks like it’s going to be a ton of fun.

Posted by dsquared · September 9, 2004 10:13 PM
Followups

This discussion has been closed. Thanks to everyone who contributed.