Email not displaying correctly? View it in your browser.
Portfolio Probe

Investment technology for the 21st century.
Version 1.03 released

A new version of Portfolio Probe has just been released.  You can update to the new version from within R (assuming no firewall issues) with the command:

install.packages("PortfolioProbe", repos="")

Below are the highlights of the changes, fuller details are in the change log:

bench.weights argument

The newly added bench.weights argument allows you to give the weights (in terms of the other assets) of one or more benchmarks.  When you do this, you don't need to add those benchmarks to the variance matrix or the expected returns -- that is done automatically.

risk.fraction argument

The risk.fraction argument allows you to constrain the fraction of variance attributable to each asset.  This allows you to create risk parity portfolios.  It also allows you to replace weight constraints with constraints on risk, which is really what they are commonly aiming at.  More details, including some mathematics, are in the recent blog post:

Random portfolio bug fix

The new version has some changes to make random portfolios closer to uniformly distributed.  In particular, assets that had a forced trade were much too likely to only trade by the forced amount.

Thank you

Thank you for your interest in Portfolio Probe.  The addition of the risk.fraction argument was user initiated.  If you have ideas of additions or changes, please let us know.

You have been sent this email because you have opted-in on our online request form.

Unsubscribe <<Email Address>> from this list.

Our mailing address is:
Burns Statistics
4-b Jodrell Road
London, England E3 2LA
United Kingdom

Add us to your address book

Copyright (C) 2011 Burns Statistics All rights reserved.

Forward this email to a friend
Update your profile
Email Marketing Powered by Mailchimp