nielsenmb
commented
3 years ago I think we did have epsilon as an optional input initially, so I don't think it'd be an issue to put it back in. It might also be a sensible thing to do when it comes to updating the prior with F-type stars, which we don't currently have a lot of.
It's just curious that for a star at ~100muHz, the prior shouldn't allow that range of epsilon at all. Also d02 should be 0.5muHz-1.25muHz or similar. That looks more like 0.25.
Did you run the fit with the sampler='cpnest'? That might be a quickfix for the issue.
On Sat, 6 Nov 2021 at 04:24, Joel Ong @.***> wrote:
In some cases, the prior placed on ε during the asymptotic fitting procedure is exactly out of phase with the true locations of the radial/quadrupole-mode ridges; however, the presence of the dipole mixed-mode forest permits a "good enough" fit to nonetheless be obtained (see attached figure: echelle.pdf https://github.com/grd349/PBjam/files/7489874/echelle.pdf). In such cases, it is often possible to demand (by eye) a small range of values for $\epsilon_0$ in the asymptotic fit, with probably better accuracy than the KDE provides. Perhaps a manual override of the ε range should be permitted?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/grd349/PBjam/issues/264, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEJWO3Z3BQ7A3HWUL7ACEETUKS3ZPANCNFSM5HPFDO7Q . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
In some cases, the prior placed on ε during the asymptotic fitting procedure is exactly out of phase with the true locations of the radial/quadrupole-mode ridges; however, the presence of the dipole mixed-mode forest permits a "good enough" fit to nonetheless be obtained (see attached figure: echelle.pdf). In such cases, it is often possible to demand (by eye) a small range of values for $\epsilon_0$ in the asymptotic fit, with probably better accuracy than the KDE provides. Perhaps a manual override of the ε range should be permitted?