timcdlucas / INLAutils

R package providing utilities for INLA
Other
26 stars 10 forks source link

Does the sloo readme make sense? #49

Closed timcdlucas closed 6 years ago

timcdlucas commented 6 years ago

Hi @pyt215 I'm just tidying up the readme and reknitting it.

https://github.com/timcdlucas/INLAutils/tree/dev

I think the code that was in there wasn't working. So I took code from the examples. But now I'm a bit confused because it looks like it has binomial response data, fitted with a normal likelihood in the first case and with a gamma in the sloo line?

I'll try and work it out, or maybe you could let me know what it should be.

pyt215 commented 6 years ago

Hi Tim,

That is very strange indeed, sorry for not having observed that. I have very quickly looked at the code but can’t see where there might be a mistake. I will now take a flight to Geneva so won’t have time to look at it before Monday (unless I have a quick look this weekend).

Andre Python

From: Tim Lucas [mailto:notifications@github.com] Sent: 18 May 2018 17:03 To: timcdlucas/INLAutils INLAutils@noreply.github.com Cc: Andre Python andre.python@bdi.ox.ac.uk; Mention mention@noreply.github.com Subject: [timcdlucas/INLAutils] Does the sloo readme make sense? (#49)

Hi @pyt215https://github.com/pyt215 I'm just tidying up the readme and reknitting it.

https://github.com/timcdlucas/INLAutils/tree/dev

I think the code that was in there wasn't working. So I took code from the examples. But now I'm a bit confused because it looks like it has binomial response data, fitted with a normal likelihood in the first case and with a gamma in the sloo line?

I'll try and work it out, or maybe you could let me know what it should be.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/timcdlucas/INLAutils/issues/49, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AUWhXPcVCY5eG11GpOm4HGPuvxTm8rgUks5tzvDGgaJpZM4UE-44.

pyt215 commented 6 years ago

Hi,

Actually in your test , you kept gamma as family

See

family <- list('gamma') #one model

you just need to replace gamma by normal in your case. Hope that’s the only issue Andre Python

From: Tim Lucas [mailto:notifications@github.com] Sent: 18 May 2018 17:03 To: timcdlucas/INLAutils INLAutils@noreply.github.com Cc: Andre Python andre.python@bdi.ox.ac.uk; Mention mention@noreply.github.com Subject: [timcdlucas/INLAutils] Does the sloo readme make sense? (#49)

Hi @pyt215https://github.com/pyt215 I'm just tidying up the readme and reknitting it.

https://github.com/timcdlucas/INLAutils/tree/dev

I think the code that was in there wasn't working. So I took code from the examples. But now I'm a bit confused because it looks like it has binomial response data, fitted with a normal likelihood in the first case and with a gamma in the sloo line?

I'll try and work it out, or maybe you could let me know what it should be.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/timcdlucas/INLAutils/issues/49, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AUWhXPcVCY5eG11GpOm4HGPuvxTm8rgUks5tzvDGgaJpZM4UE-44.

timcdlucas commented 6 years ago

Ok, I'll just edit it do it makes sense. Don't get round to it till next week though.