Release of Portfolio Probe version 1.05 plus blog problems

June 25, 2012

June has been a good month for releases of Portfolio Probe and a bad month for disseminating the blog.

Version 1.05

This is a minor release, primarily to fix some bugs, but a little sugar slipped in as well.

  • bug: when forced trades created a portfolio that satisfied all of the constraints, then all “random” portfolios were the same.
  • bug: constraints.realized now (again) properly handles additional constraints that were not initially imposed.
  • bug: some problems wasted a lot of time.  The key ingredients were a variance and trading no more than 25 assets.
  • convenience: the risk.fraction argument can now take a list.
  • convenience: the build.constraints function can now take a list.

More detail is in the change log and especially in the user support email.

Blog problems

For almost all of June it seems that the email feed for the blog has not been working.  Correct me if I’m wrong.

Also the R Language RSS feed seems not to have been working either.  This means that the posts involving R have not been appearing on R-bloggers or statsblogs.

I’ve not seen any problems with the main RSS feed.

By all appearances Google is not supporting Feedburner very well as this blog doesn’t seem to be an isolated case.  Suggestions are welcome.

 

Leave a Reply

Related posts

  • May 15, 2012

    There have been some issues with the sign-up process for the Portfolio Probe newsletter and the Portfolio Probe user's list.  The issues may or may not be in the [...]

  • March 28, 2012

    How much turnover is required to get portfolios back to their constraints? Previously “Low (and high) volatility strategy effects” created 6 sets of random portfolios as of 2007 and [...]

  • October 13, 2011

    A different sort of generalization of variance partitions. Previously The post "Generalizing risk fractions" described additional (to version 1.04 of Portfolio Probe) ways of dividing the variance among the [...]