Huge outliers about animotum

[otoolem]

Hi Ian. Yes that was very helpful. All sorted now. Thanks Malcolm O’Toole UWA Research Associate IOMRC building Room 2.43 Cnr Fairway and Service Road 4 (M470) The University of Western Australia 35 Stirling Highway, Crawley Western Australia 6009 Phone: [phone] Mobile: [phone] Email: [email protected]

…

________________________________ From: Ian Jonsen <[email protected]> Sent: Friday, May 10, 2019 18:08 To: ianjonsen/foieGras Cc: Malcolm O'Toole; Author Subject: Re: [ianjonsen/foieGras] Huge outliers (#5) Hi Malcolm, given your recent email re: outlier estimates no longer occurring, are you happy for me to close this issue? thanks, Ian — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#5 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ABKEWPAGBWDZHI3VAM7QT4DPUVCQ5ANCNFSM4HEGALCA>. University of Tasmania Electronic Communications Policy (December, 2014). This email is confidential, and is for the intended recipient only. Access, disclosure, copying, distribution, or reliance on any of it by anyone outside the intended recipient organisation is prohibited and may be a criminal offence. Please delete if obtained in error and email confirmation to the sender. The views expressed in this email are not necessarily the views of the University of Tasmania, unless clearly intended otherwise.

[ianjonsen]

Hmm, I'm not sure, Malcolm. Can you give me a reproducible example, ie. enough data to fit the model?

[otoolem]

Hi Ian I have tried to re-run fit_ssm to investigate the problem of outliers but it no longer gives me the correct output. If I apply the speed filter I receive a warning message: Unknown or uninitialised column: 'data'. If I apply the crw model I get output without predicted lon/lat columns (fit$data has no lon/lat columns). I’ve attached some data, and the code I use is below. fit1=fit_ssm(d, vmax=3, time.step=24, pf=TRUE) fit2=fit_ssm(d, vmax=3, time.step=24, model='crw') I have tried running foieGras on different versions of R and still no luck. I wonder if I am missing crucial packages (I re-installed TMB and Matrix and still nothing). Any thoughts? Cheers Malcolm Malcolm O’Toole UWA Research Associate IOMRC building Room 2.45 1 Cnr Fairway and Service Road 4 (M470) The University of Western Australia 35 Stirling Highway, Crawley Western Australia 6009 Phone: [phone] Mobile: [phone] Email: [email protected]

…

________________________________ From: Ian Jonsen <[email protected]> Sent: Monday, April 8, 2019 9:18:25 PM To: ianjonsen/foieGras Cc: Malcolm O'Toole; Author Subject: Re: [ianjonsen/foieGras] Huge outliers (#5) Hmm, I'm not sure, Malcolm. Can you give me a reproducible example, ie. enough data to fit the model? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#5 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AFRLPGA0svp6pFnlyzSpBOAYn0BclNeHks5ve0GhgaJpZM4chgLE>. University of Tasmania Electronic Communications Policy (December, 2014). This email is confidential, and is for the intended recipient only. Access, disclosure, copying, distribution, or reliance on any of it by anyone outside the intended recipient organisation is prohibited and may be a criminal offence. Please delete if obtained in error and email confirmation to the sender. The views expressed in this email are not necessarily the views of the University of Tasmania, unless clearly intended otherwise.

[ianjonsen]

It's not you, it's me. think I broke prefilter when pf = TRUE as I got the same error late yesterday. I'll work on a fix and then we can get back to the issue of the strange estimates

[otoolem]

Thanks Ian. Much appreciated. Cheers Malcolm Malcolm O’Toole UWA Research Associate IOMRC building Room 2.43 Cnr Fairway and Service Road 4 (M470) The University of Western Australia 35 Stirling Highway, Crawley Western Australia 6009 Phone: [phone] Mobile: [phone] Email: [email protected]

…

________________________________ From: Ian Jonsen <[email protected]> Sent: Wednesday, April 10, 2019 19:56 To: ianjonsen/foieGras Cc: Malcolm O'Toole; Author Subject: Re: [ianjonsen/foieGras] Huge outliers (#5) It's not you, it's me. think I broke prefilter when pf = TRUE as I got the same error late yesterday. I'll work on a fix and then we can get back to the issue of the strange estimates — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#5 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AFRLPD9VGQAsd6z889sene7tgIUBd2oGks5vfdFWgaJpZM4chgLE>. University of Tasmania Electronic Communications Policy (December, 2014). This email is confidential, and is for the intended recipient only. Access, disclosure, copying, distribution, or reliance on any of it by anyone outside the intended recipient organisation is prohibited and may be a criminal offence. Please delete if obtained in error and email confirmation to the sender. The views expressed in this email are not necessarily the views of the University of Tasmania, unless clearly intended otherwise.

[ianjonsen]

Hi Malcolm, given your recent email re: outlier estimates no longer occurring, are you happy for me to close this issue? thanks, Ian