<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Here's some background reading from April's edition of the Journal
    of Finance.  This is slightly off topic, but I had been intending to
    send this paper to this group and this reminded me that I should
    send it around.  The conclusion: For portfolios to perform well out
    of sample, it is important to incorporate an ensemble of methods
    that cover stochastic volatility, time-varying return expectation,
    AND parameter uncertainty.  It would be great to take this paper as
    an excuse to start to develop best practices for some of these
    issues.<br>
    <br>
    pcc<br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 05/19/2014 10:20 AM, Ross Bennett
      wrote:<br>
    </div>
    <blockquote
cite="mid:CAEESW=XVB2yOYxifEe-HXmp-YFBTdNU01HAHNwdHCMD8YtD+ug@mail.gmail.com"
      type="cite">
      <div dir="ltr"><font face="arial, helvetica, sans-serif">Mentors
          (and others on the list),</font>
        <div><font face="arial, helvetica, sans-serif"><br>
          </font></div>
        <div><font face="arial, helvetica, sans-serif">It was great
            seeing you all at R/Finance last week.</font></div>
        <div><font face="arial, helvetica, sans-serif"><br>
          </font></div>
        <div><font face="arial, helvetica, sans-serif">Here are the
            first 3 tasks in the timeline in my original proposal.</font></div>
        <div>
          <p style="color:rgb(0,0,0)">
            <font face="arial, helvetica, sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">5/19
                - 5/30: </span><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">Multilayer
                Optimization</span></font></p>
          <p style="color:rgb(0,0,0)"><font face="arial, helvetica,
              sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">5/31
                - 6/13: </span><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">Regime
                Switching Optimization</span></font></p>
          <p style="color:rgb(0,0,0)"><font face="arial, helvetica,
              sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">6/14
                - 6/27: </span><span
                style="line-height:14px;white-space:pre-wrap">Factor
                Portfolio Optimization</span></font></p>
          <p style="color:rgb(0,0,0)"><span
              style="font-family:arial,helvetica,sans-serif"><br>
            </span></p>
          <p style="color:rgb(0,0,0)"><span
              style="font-family:arial,helvetica,sans-serif">I would
              like to modify the timeline as follows:</span><br>
          </p>
          <p style="color:rgb(0,0,0)"><font face="arial, helvetica,
              sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">5/19
                - 5/30: </span></font><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">Regime
              Switching Optimization</span></p>
          <p style="color:rgb(0,0,0)"><font face="arial, helvetica,
              sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">5/31
                - 6/13: </span></font><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">Factor
              Portfolio Optimization</span></p>
          <p style="color:rgb(0,0,0)"><font face="arial, helvetica,
              sans-serif"><span
style="background-color:transparent;line-height:1.15;white-space:pre-wrap">6/14
                - 6/27: </span></font><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">Multilayer
              Optimization</span></p>
          <p style="color:rgb(0,0,0)"><br>
          </p>
          <p style="color:rgb(0,0,0)"><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">Regime
              Switching Optimization</span></p>
          <p><font color="#000000" face="arial, helvetica, sans-serif"><span
                style="line-height:14px;white-space:pre-wrap">My code
                example for the proposal was for regime switching. I
                kept it pretty general so that the user can just pass in
                a time series of what portfolio to use depending on the
                regime for any arbitrary regime switching model. Do you
                have any comments or feedback on my proposed way for
                supporting regime switching?</span></font></p>
          <p style="color:rgb(0,0,0)"><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">The
              regime switching example is in the following code gist</span></p>
          <p style="color:rgb(0,0,0)"><span style="color:rgb(34,34,34)"><a
                moz-do-not-send="true"
                href="https://gist.github.com/rossb34/9645489">https://gist.github.com/rossb34/9645489</a></span></p>
          <p style="color:rgb(0,0,0)"><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">
            </span></p>
          <p style="color:rgb(0,0,0)"><span
style="font-family:arial,helvetica,sans-serif;line-height:14px;white-space:pre-wrap">Factor
              Models to Estimate Moments</span></p>
          <p><span
              style="color:rgb(0,0,0);font-family:arial,helvetica,sans-serif">I
              reached out to Kris Boudt a few weeks ago and he shared
              some of the code he used for his paper. I put together a
              small package as a prototype. One thing to note is that
              Kris sent me C++ code (using Rcpp) for computing the
              cokurtosis error matrix. I would prefer not to rewrite the
              code in R, but this does mean introducing a dependency on
              Rcpp.  Along these same lines, we could also write
              functions to compute the 3rd and 4th moments in C++. The
              Armadillo library has a k</span><font color="#000000"
              face="arial, helvetica, sans-serif">ron() function to
              compute the kronecker product making the higher moments an
              easy computation using RcppArmadillo. We could do this as
              a prototype in PortfolioAnalytics and eventually move to
              PerformanceAnalytics so <a moz-do-not-send="true"
                href="http://M3.MM">M3.MM</a> and <a
                moz-do-not-send="true" href="http://M4.MM">M4.MM</a> are
              compiled code. Thoughts?</font><br>
          </p>
          <p><font color="#000000" face="arial, helvetica, sans-serif">Here
              is the source code of the prototype package for moment
              estimates.</font></p>
          <p><a moz-do-not-send="true"
              href="https://bitbucket.org/rossbennett34/momentestimation/src">https://bitbucket.org/rossbennett34/momentestimation/src</a></p>
          <p><br>
          </p>
          <p>Best,</p>
          <p>Ross</p>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
GSoC-PortA mailing list
<a class="moz-txt-link-abbreviated" href="mailto:GSoC-PortA@lists.r-forge.r-project.org">GSoC-PortA@lists.r-forge.r-project.org</a>
<a class="moz-txt-link-freetext" href="http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta">http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>