When the money keeps rolling out you don't keep books You can tell you've done well by the  grateful looks Accountants only slow things down, figures get in the way... “And the Money Kept Rolling In (and Out)”
Evita, The Motion Picture
I received an email the week of April 20 from Mike Driscoll, co-chair (with Jim Porzak) of the Bay Area R Users Group, announcing the formation of three new such groups – Los Angeles, New York, and Ottawa – and citing their inaugural meetings. Driscoll also noted a request to have the burgeoning user group list included on the R project page, for me yet another indicator of the explosive growth of an organizing R community.
There's no official Chicago R user group at this time, but rather an informal Chicago R finance  collaboration promoted by the Finance Department of the University of Illinois Chicago (UIC),  The International Center for Futures and Derivatives (ICFD) at UIC, and the Chicago R finance community.  On April 24th and 25th, this group sponsored  R/Finance 2009: Applied Finance with R, at UIC. Along with several hundred others, I had the good fortune to attend and learn from an international cast of R finance luminaries that included Patrick Burns, David Rupert, Roger Koenker, Diethelm Wuertz, Eric Zivot, Robert Grossman, and David Kane, as well as several leading  computational  finance practitioners and many R special interest group (R-SIG-Finance) contributors. 
A quantitative finance (QF) hobbyist, I'm unfortunately no more than an advanced novice in the discipline. Much of the  little I know though was gleaned from David Ruppert's excellent text, Statistics and Finance, An Introduction, the content of which the author has used to train generations of budding financial engineers at Cornell University. In a  revision of the text currently in progress, Ruppert tellingly switches form SAS to R  to illustrate his statistical techniques. Ruppert's Cornell students have made the transition from SAS to R with him, and are now R advocates in the commercial world. 
Diethelm Wuertz and Guy Yollin presented complimentary R tools for  portfolio optimization. Wuertz, an econophysicist (yikes!) at ETH, Zurich, is primary author of a comprehensive and now mature computational finance R package  Rmetrics.   An accomplished academic, Wuertz has invested considerably of his own and student efforts for Rmetrics. His open source largesse is rewarded  with commercial workshops and ebooks detailing his methods.  Yolllin of Rotella Capital Management, on the other hand, is less grand , citing Picasso: “Good Artists Borrow. Great Artists Steal.” He's also an excellent presenter. Yollin's exposition on using R for portfolio optimization resounded even with me; his supporting R code was lucid and didactic.
Economist Roger Koenker presented on Quantile Regression for Fin and Fun. In contrast to traditional least squares regression, which revolves on the conditional mean of the dependent (y) variable, quantile regression (QR) models conditional quantile functions or percentiles of y. I like QR a lot, using it often in my analytic work, especially for determining if predictive behavior at the extremes of  distributions is different from that of the means. Koenker's R quantile regression package is an important contribution to predictive modeling for BI.
Performance Measurement (PM) is a pertinent topic for finance and investments, just as it's central to business intelligence.  Investment  PM is concerned with portfolio returns and risk, as well as with the attribution of efficiency and skill to managers. Too often, specific portfolio performance is compared to simple indexes like the SP 500 and the Russell 2000 that are generally inadequate for the task. Patrick Burns and David Kane presented two complimentary methods for assessing investment manager skill.  Burns proposed comparing the manager's work with a benchmark random portfolio with the same allocation constraints, while Kane proffered a portfolio matched to the manager's on exposures but with different components.  The manager's actual returns contrasted with those of the random or matched portfolios provide a sound basis for assessing skill, thus showing investors what they're getting for their often considerable fees. Coincidentally, randomization and matching are also two important methods to help businesses evaluate the performance of their strategic initiatives, and should be staples of  the BI designer' s tool chest.
Kudos to Gib Bassett, who heads both the UIC Finance Department and the ICFD, other UIC faculty and staff, the outstanding presenters, and the many talented Chicago-area QF practitioners and R-SIG-Finance contributors for making this conference a rousing success.
Steve Miller's blog can also be found at miller.openbi.com.