LondonR recap

The biggest and perhaps best meeting yet.

The talks

James Long: “Easy Parallel Stochastic Simulations using Amazon’s EC2 & Segue”.  This was a lively talk about James’ package to use Amazon’s cloud to speed up a (huge) call to lapply.  The good part is that if you want to use Amazon as your cloud provider, it is very easy.  The downside is that the package is intimately tied to Amazon — so if you want a more general solution, you should probably go elsewhere.

Richard Saldanha: “Practical Optimization”.  In this Richard is essentially generating random portfolios in order to understand the capacity (amount of money possible) of a strategy.

Chibisi Chima-Okereke: “Actuarial Pricing Using General Linear Models In R”. Chibisi talked about how R has not yet had much penetration into the actuarial field, and showed some things that they are missing because they don’t have R.

The good part

The crowd was significantly bigger than at past events, and seemed to be much more diverse as well.  I met several quite interesting people, and I’m sure there were many quite interesting people that I didn’t get to meet.

Why would it be a bigger crowd?  There are at least three hypotheses:

1. Food was served.
2. An out-of-towner in the form of James Long presented.
3. R is growing like a weed.

I’m not at all convinced of hypothesis 1.  The additional attraction of free protein beyond the attraction of free ethanol would seem to me to be marginal.  Plus I wasn’t even consciously aware that there would be food (but it was good — thanks Mango).

I would guess that James did attract some people — and rightly so.

Most of the weight of my prior is on hypothesis 3.  There have been a few books published recently and other events that may well be raising awareness of R:

• The Art of R Programming by Norman Matloff was published 2011 September
• R in Action by Robert Kabacoff was published 2011 August
• Parallel R by Ethan McCallum and Stephen Weston was published 2011 October (I certainly wouldn’t have predicted this one)
• Last week a new website was born: http://rconvert.com/ (I’m not saying this had an effect on the meeting, but it’s a sign of the times)

Other evidence in support of hypothesis 3 is that views of “Some hints for the R beginner” have been growing, and downloads of The R Inferno remain strong.

The next data point regarding the growth of LondonR is scheduled for 2012 March 6.  A call went out for suggestions for a larger suitable venue in the City — talk to Mango Solutions if you have an idea.

Another view of the same meeting is on Mages’ blog

Subscribe to the Portfolio Probe blog by Email