CosmoCoffee

Hi Antony --

I am interested in producing gridded likelihoods for three parameters, i.e. the equivalent of the <file_root>_2D_x_y_likes files produced by getdist, except for three parameters instead of two.

Any easy way to do this?

-- Will

Will Kinney wrote:Hi Antony --

I am interested in producing gridded likelihoods for three parameters, i.e. the equivalent of the <file_root>_2D_x_y_likes files produced by getdist, except for three parameters instead of two.

Any easy way to do this?

This should have had "COSMOMC" in the subject line!

Looks relatively straightforward to hack a new subroutine for getdist to accomplish this, but there must be a more elegant solution...

-- Will

It is not in GetDist, but I think you can do it using CosmoloGui or the matlab scripts that that uses (e.g. see the 3D contour plots). Sarah may be able to help further here.

Hi, yes, the relevant routines are in the cosmomc/mscripts directory (cosmologui)
e.g. could call it like this:
[xim,xn]=samp3imw([x' y' z']',samples,resolution); % grid
handle=plot3dconf(xn(1,:),xn(2,:),xn(3,:),xim,0.68,5,1); % isosurface

Sarah Bridle wrote:Hi, yes, the relevant routines are in the cosmomc/mscripts directory (cosmologui)
e.g. could call it like this:
[xim,xn]=samp3imw([x' y' z']',samples,resolution); % grid
handle=plot3dconf(xn(1,:),xn(2,:),xn(3,:),xim,0.68,5,1); % isosurface

Thanks much!

-- Will