R help - Feb 2012 - 3D confidence ellipsoid with ellipse projections onto 2D plane

I have a 3xN matrix of parameters obtained from N regressions where the 3
parameters are jointly statistically significant. I would like to reproduce
a 3D confidence ellipsoid projecting 2D ellipses onto the XY plane as in
Figure 5.2 in this
text<http://books.google.com/books?id=PMeJGeXA09EC&pg=PA172&dq=confidence+ellipsoid&hl=en&sa=X&ei=oWQxT4OgBYLSiALW85WtAw&ved=0CEIQ6AEwAg#v=onepage&q=confidence%20ellipsoid&f=false>
.

Is this possible using some combination of ellipse3d() and ellipse()? Any
insights would be greatly appreciated.

Thanks.

	[[alternative HTML version deleted]]

Hi,

If you read French, you might find the following discussion interesting,

http://www.forum.math.ulg.ac.be/viewthread.html?id=45765

It contains some good suggestions to project an ellipsoid onto a
plane, which as I understand might be related to your question.

HTH,

b.



On 8 February 2012 07:29, Samuel Bazzi <sbazzi at ucsd.edu>
wrote:
> I have a 3xN matrix of parameters obtained from N regressions where the 3
> parameters are jointly statistically significant. I would like to reproduce
> a 3D confidence ellipsoid projecting 2D ellipses onto the XY plane as in
> Figure 5.2 in this
>
text<http://books.google.com/books?id=PMeJGeXA09EC&pg=PA172&dq=confidence+ellipsoid&hl=en&sa=X&ei=oWQxT4OgBYLSiALW85WtAw&ved=0CEIQ6AEwAg#v=onepage&q=confidence%20ellipsoid&f=false>
> .
>
> Is this possible using some combination of ellipse3d() and ellipse()? Any
> insights would be greatly appreciated.
>
> Thanks.
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

On 2/7/2012 1:29 PM, Samuel Bazzi wrote:
> I have a 3xN matrix of parameters obtained from N regressions where the 3
> parameters are jointly statistically significant. I would like to reproduce
> a 3D confidence ellipsoid projecting 2D ellipses onto the XY plane as in
> Figure 5.2 in this
>
text<http://books.google.com/books?id=PMeJGeXA09EC&pg=PA172&dq=confidence+ellipsoid&hl=en&sa=X&ei=oWQxT4OgBYLSiALW85WtAw&ved=0CEIQ6AEwAg#v=onepage&q=confidence%20ellipsoid&f=false>
> .
>
> Is this possible using some combination of ellipse3d() and ellipse()? Any
> insights would be greatly appreciated.
>
The situation described in Rawlings is for a single regression model, 
with an associated covariance matrix of the parameters which is the 
basis for the ellipsoid.
If all you have are the coefficient estimates, where do you think your
ellipsoid can come from?

Nevertheless, the situation you described -- plotting confidence 
ellipsoids for a model with projections on the XY and other planes
is handled by the p3d package, only on R-Forge
https://r-forge.r-project.org/projects/p3d/
It is in alpha development, and not completely documented.

HTH
-Michael


-- 
Michael Friendly     Email: friendly AT yorku DOT ca
Professor, Psychology Dept.
York University      Voice: 416 736-5115 x66249 Fax: 416 736-5814
4700 Keele Street    Web:   http://www.datavis.ca
Toronto, ONT  M3J 1P3 CANADA