Access to PSDs?

[avivajpeyi]

Hi @jroulet!

I tried plotting the LVC GW150914 posterior samples (obtained from GWOSC) against your GW150914 samples, and the results appear rather different:

I was wondering -- could you upload your PSDs that you used for your analysis? Id like to try re-doing your analysis to better understand why the results are so different.

Thanks!

[jroulet]

Dear Avi, Thank you for your interest. I do not recover your plot when I download the samples myself from the repository, please find attached the plot I get. The culprits seem to be the transformations of the mass parameters. The samples report `mchirp` and `eta`. To be sure: `mchirp` is the detector-frame chirp mass, i.e. mchirp = (1+z) * (m1_source*m2_source)**0.6 / (m1_source+m2_source)**0.2 where z is the cosmological redshift and m1_source, m2_source are the source-frame component masses. `eta` is the symmetric mass ratio: eta = m1*m2/(m1+m2)**2 Could you please double-check whether you are using these definitions? This said, if you still would like to have the PSDs let me know and I will be happy to provide them. Best regards, Javier El lun, 30 nov 2020 a las 21:31, Avi Vajpeyi (<notifications@github.com>) escribió:

…

Hi @jroulet <https://github.com/jroulet>! I tried plotting the LVC GW150914 posterior samples (obtained from GWOSC) against your GW150914 samples, and the results appear rather different: [image: test] <https://user-images.githubusercontent.com/15642823/100689792-4e544f00-33d9-11eb-9c49-eb7ee08ce2c2.png> I was wondering -- could you upload your PSDs that you used for your analysis? Id like to try re-doing your analysis to better understand why the results are so different. Thanks! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEMTR4V3WUJ4X44W7WTZ5Z3SSRIQTANCNFSM4UIMKVRQ> .

[avivajpeyi]

Ah, I did not realise the samples headings were for

• "eta": "mass ratio" --> "symmetric mass ratio"
  

I interpreted them incorrectly!

Correcting the above, I get:

(haven't calculated the z values as they take a while to compute).

---

Would it still be possible to upload your PSDs?