There are several new things in the Portfolio Probe world.

There is a second beta available of Portfolio Probe version 1.04.  The two main changes are:

• It works under R 2.14.x (a namespace was added)
• Specification of trading costs is generalized (by user request)

BurStFin is now on CRAN

The BurStFin R package is (finally) on CRAN.  It is almost the same as the old version, but has a bit of improvement:

• It works in R 2.14.x (a namespace was added)
• The variance estimation functions (Ledoit-Wolf shrinkage and statistical factor model) now warn by default if you accidentally give them prices rather than returns
• There is a new function that creates a three-dimensional array from one or more matrices

Using Portfolio Probe in C++

The real work of random portfolio generation and trade optimization in Portfolio Probe is done in C.  So it would seem that if you want to use Portfolio Probe in a C++ program that calling that code directly would be the thing to do.  In actuality that is a very hard path to go down.

But there's an alternative.  Thanks to the wonderful R packages Rcpp and RInside it is quite easy to call R code from C++.  This includes Portfolio Probe.

There is now an example of doing this on the Cookbook page on the Portfolio Probe website: http://www.portfolioprobe.com/user-area/documentation/portfolio-probe-cookbook/c-and-portfolio-probe-via-rinside/

Blog highlights 

The most popular blog posts in January are listed in: http://www.portfolioprobe.com/2012/02/01/popular-posts-2012-january/

The winner was the post "The top 7 portfolio optimization problems": http://www.portfolioprobe.com/2012/01/05/the-top-7-portfolio-optimization-problems/