Probit ordered specification

[DjakissO]

Hi, I want to fit a Bayesian dynamic ordered panel probit. I've read the Tikka and Helske （2023） guide but haven't seen how to specify an ordered probit. So I would like to know if I can use dynamite to specify the ordinal nature of my dependent variable, and if so, how. Many thanks for considering my request.

[santikka]

Hi @DjakissO ! Unfortunately, ordered probit is not currently supported, but we have plans to add this feature (probably soon), especially if there is interest in this feature.

[DjakissO]

Hi @@.***>,

Thank you for your feedback. I want to know if there is another Bayesian approach package that can support panel ordered probit.

Best regards, Djakalia

---

From: Santtu Tikka @.> Sent: Monday, April 8, 2024 2:00 PM To: ropensci/dynamite @.> Cc: DjakissO @.>; Mention @.> Subject: Re: [ropensci/dynamite] Probit ordered specification (Issue #79)

Hi @DjakissOhttps://github.com/DjakissO ! Unfortunately, ordered probit is not currently supported, but we have plans to add this feature (probably soon), especially if there is interest in this feature.

— Reply to this email directly, view it on GitHubhttps://github.com/ropensci/dynamite/issues/79#issuecomment-2042837042, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BHUSPJS6VMJJBIQAWMDR6IDY4KPH5AVCNFSM6AAAAABF3AC7GGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANBSHAZTOMBUGI. You are receiving this because you were mentioned.Message ID: @.***>

[santikka]

@DjakissO Ordered probit and logit are now available in dynamite, you can write, e.g., obs(y ~ x, family = "cumulative", link = "probit") The response for such a channel should be an ordered factor in the data. Please keep in mind that this feature is not yet extensively tested.

[DjakissO]

Hi,

Wow! That's great news. Thank you very much for the information.

All the best, Djakalia

Le mer. 24 avr. 2024, 14:52, Santtu Tikka @.***> a écrit :

@DjakissO https://github.com/DjakissO Ordered probit and logit are now available in dynamite, you can write, e.g., obs(y ~ x, family = "cumulative", link = "probit") The response for such a channel should be an ordered factor in the data. Please keep in mind that this feature is not yet extensively tested.

— Reply to this email directly, view it on GitHub https://github.com/ropensci/dynamite/issues/79#issuecomment-2075138848, or unsubscribe https://github.com/notifications/unsubscribe-auth/BHUSPJVYBBBDHAP3ERERRODY67BJ7AVCNFSM6AAAAABF3AC7GGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZVGEZTQOBUHA . You are receiving this because you were mentioned.Message ID: @.***>

[DjakissO]

Hi @santikka, I tried to run it but ti returns the following error message: unused argument (link = "probit"). What can I do to resolve this error ?

[santikka]

Hi @DjakissO, please install the latest development version of dynamite, as this update is not yet on CRAN: devtools::install_github("ropensci/dynamite")

[DjakissO]

It's perfect now. Thank you for your help, and especially for your diligence.

---

From: Santtu Tikka @.> Sent: Thursday, April 25, 2024 5:05 AM To: ropensci/dynamite @.> Cc: DjakissO @.>; Mention @.> Subject: Re: [ropensci/dynamite] Probit ordered specification (Issue #79)

Hi @DjakissOhttps://github.com/DjakissO, please install the latest development version of dynamite, as this update is not yet on CRAN: devtools::install_github("ropensci/dynamite")

— Reply to this email directly, view it on GitHubhttps://github.com/ropensci/dynamite/issues/79#issuecomment-2076363370, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BHUSPJXEGPTXTWPES7KJGA3Y7CFKNAVCNFSM6AAAAABF3AC7GGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZWGM3DGMZXGA. You are receiving this because you were mentioned.Message ID: @.***>

[santikka]

Closing this as completed.