<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=us-ascii"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.MsoPlainText, li.MsoPlainText, div.MsoPlainText
{mso-style-priority:99;
mso-style-link:"Plain Text Char";
margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
span.PlainTextChar
{mso-style-name:"Plain Text Char";
mso-style-priority:99;
mso-style-link:"Plain Text";
font-family:"Calibri","sans-serif";}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]--></head><body lang=EN-US link=blue vlink=purple><div class=WordSection1><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>-----Original Message-----<br>From: gsoc-porta-bounces@lists.r-forge.r-project.org [mailto:gsoc-porta-bounces@lists.r-forge.r-project.org] On Behalf Of Brian G. Peterson<br>Sent: Sunday, October 06, 2013 6:23 AM<br>To: gsoc-porta@r-forge.wu-wien.ac.at<br>Subject: Re: [GSoC-PortA] Mean-mETL objective?</p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Doug,<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>If the underlying solvers can do it, we should be able to pass it through to ROI. There shouldn't be any inherent restriction from using ROI.<o:p></o:p></p><p class=MsoPlainText><b><i><span style='color:black'>[Doug] Agreed.</span></i></b><span style='color:black'><o:p></o:p></span></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>I think, based on reading your replies, that we just need to understand how to formulate the problem.<o:p></o:p></p><p class=MsoPlainText><b><i><span style='color:black'>[Doug] That is done.</span></i></b><span style='color:black'><o:p></o:p></span></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>DEoptim only finds one minima, and it appears to be the correct one. <o:p></o:p></p><p class=MsoPlainText>'Neighbor' portfolios (the portfolios closest to the global minima by objective vaule) are all close in the hull space to the global minima.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Overall, I'd hope to investigate this a bit more, because if the formulation as a LP/QP problem is possible, it's obviously much faster than using random portfolios or DEoptim (or PSO, or GenSA, etc) to find the objective.<o:p></o:p></p><p class=MsoPlainText><b><i><span style='color:black'>[Doug] I think spend a little time on this later this autumn or early winter.</span></i></b><span style='color:black'><o:p></o:p></span></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Regards,<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Brian<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>On 10/05/2013 10:15 AM, Doug Martin wrote:<o:p></o:p></p><p class=MsoPlainText>> *Ross Bennett wrote:<o:p></o:p></p><p class=MsoPlainText>> Peter,<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> Unfortunately, with ROI we are only able to minimize ETL with ETL as <o:p></o:p></p><p class=MsoPlainText>> an objective. If you have mean and ETL as an objective using ROI, <o:p></o:p></p><p class=MsoPlainText>> unless there is a target in the mean return objective, we just <o:p></o:p></p><p class=MsoPlainText>> minimize ETL. If you have both mean and ETL as objectives, you could <o:p></o:p></p><p class=MsoPlainText>> add a target to the mean objective and this will minimize ETL subject to the target return.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> */[Doug] That is unfortunate./*<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> We can do the following with ETL as an objective using ROI:<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> - Minimize ETL subject to leverage, box, group, exposure, position <o:p></o:p></p><p class=MsoPlainText>> limit, and target return.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> Multipliers are ignored with ROI since the problems are formulated <o:p></o:p></p><p class=MsoPlainText>> into an LP/QP problem.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> */[Doug] Ditto on my comment above. There is no problem to use <o:p></o:p></p><p class=MsoPlainText>> solve.QP directly in the quadratic utility (QU) formulation of the <o:p></o:p></p><p class=MsoPlainText>> problem (as well as the MinVar formulation). Should be the same for <o:p></o:p></p><p class=MsoPlainText>> Rglpk. So we are inheriting a limitation of ROI not of the solvers <o:p></o:p></p><p class=MsoPlainText>> themselves./*<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> *//*<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> I'll take a look at the documentation in optimize.portfolio and make <o:p></o:p></p><p class=MsoPlainText>> sure this is clear.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> I hope that helps clear it up.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> Ross<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> On Fri, Oct 4, 2013 at 11:49 AM, Peter Carl <peter@braverock.com <o:p></o:p></p><p class=MsoPlainText>> <<a href="mailto:peter@braverock.com"><span style='color:windowtext;text-decoration:none'>mailto:peter@braverock.com</span></a>>> wrote:<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> Hey Ross,<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> I can't seem to get the Mean-mETL objective to select anything other <o:p></o:p></p><p class=MsoPlainText>> than the Min mETL portfolio using ROI. It looks like there should be <o:p></o:p></p><p class=MsoPlainText>> good convexity, but I think there's a substantial imbalance between <o:p></o:p></p><p class=MsoPlainText>> the size of the monthly mean return and the loss indicated by the ETL. <o:p></o:p></p><p class=MsoPlainText>> I've tried modifying the multiplier on the mean, but it doesn't seem <o:p></o:p></p><p class=MsoPlainText>> to have an effect.<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> Any thoughts?<o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> pcc<o:p></o:p></p><p class=MsoPlainText>> --<o:p></o:p></p><p class=MsoPlainText>> Peter Carl<o:p></o:p></p><p class=MsoPlainText>> <a href="http://www.braverock.com/peter"><span style='color:windowtext;text-decoration:none'>http://www.braverock.com/peter</span></a><o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> _______________________________________________<o:p></o:p></p><p class=MsoPlainText>> GSoC-PortA mailing list<o:p></o:p></p><p class=MsoPlainText>> <a href="mailto:GSoC-PortA@lists.r-forge.r-project.org"><span style='color:windowtext;text-decoration:none'>GSoC-PortA@lists.r-forge.r-project.org</span></a><o:p></o:p></p><p class=MsoPlainText>> <<a href="mailto:GSoC-PortA@lists.r-forge.r-project.org"><span style='color:windowtext;text-decoration:none'>mailto:GSoC-PortA@lists.r-forge.r-project.org</span></a>><o:p></o:p></p><p class=MsoPlainText>> <a href="http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta"><span style='color:windowtext;text-decoration:none'>http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta</span></a><o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText>> _______________________________________________<o:p></o:p></p><p class=MsoPlainText>> GSoC-PortA mailing list<o:p></o:p></p><p class=MsoPlainText>> <a href="mailto:GSoC-PortA@lists.r-forge.r-project.org"><span style='color:windowtext;text-decoration:none'>GSoC-PortA@lists.r-forge.r-project.org</span></a><o:p></o:p></p><p class=MsoPlainText>> <a href="http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta"><span style='color:windowtext;text-decoration:none'>http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta</span></a><o:p></o:p></p><p class=MsoPlainText>><o:p> </o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>--<o:p></o:p></p><p class=MsoPlainText>Brian G. Peterson<o:p></o:p></p><p class=MsoPlainText><a href="http://braverock.com/brian/"><span style='color:windowtext;text-decoration:none'>http://braverock.com/brian/</span></a><o:p></o:p></p><p class=MsoPlainText>Ph: 773-459-4973<o:p></o:p></p><p class=MsoPlainText>IM: bgpbraverock<o:p></o:p></p><p class=MsoPlainText>_______________________________________________<o:p></o:p></p><p class=MsoPlainText>GSoC-PortA mailing list<o:p></o:p></p><p class=MsoPlainText><a href="mailto:GSoC-PortA@lists.r-forge.r-project.org"><span style='color:windowtext;text-decoration:none'>GSoC-PortA@lists.r-forge.r-project.org</span></a><o:p></o:p></p><p class=MsoPlainText><a href="http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta"><span style='color:windowtext;text-decoration:none'>http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/gsoc-porta</span></a><o:p></o:p></p></div></body></html>