Uncertainty of Fitted Parameters are the same

[Efrain]

Hello everyone,

I have been trying to fit data to a transit light curve. I have some parameters that pretty reasonable to acceptable values. But I noticed that the uncertainty on the fitted parameters by emcee seem to be the same. I am guessing it has something to do with the number of the number of walkers(2000) that I am using. I'm making sure that it is much greater than the number of dimensions (7 in this case), and I am using 5000 iterations. My corner plots seem well converged, but again, the only problem that I am seeing is uncertainty (1 sigma) for the fitted parameters is the same and very small.

Thank you for the help!

[Dan Foreman-Mackey]

Hi, if you email me directly with more details, I'll see what I can do to help. 

In general you'll want to use many fewer walkers (order 10s, not 1000s) since burn in will be long with that many walkers. 5000 iterations is probably not enough to even be burned in. You'll always want to check convergence using the techniques explained in the docs!

Best,

Dan

--

You received this message because you are subscribed to the Google Groups "emcee users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to emcee-users...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/emcee-users/98fca051-7759-4578-bb91-1733660a3e6fn%40googlegroups.com.

[Efrain Alvarado]

Thank you! That was a helpful hint to what was going on! Turns out that my log_prior function ranges were too strict. That fixed a lot of issues that I was having. I was able to decrease the number of walkers and iterations by a lot by doing so. It also fixed the error ' Too few points to create valid contours' when plotting the corner plots. Another problem that was pulling up was the plots for the position of each walker as a function of the number of steps in the chain. For me, it looked like strips instead of zig-zagged lines.