bayestestR icon indicating copy to clipboard operation
bayestestR copied to clipboard

How to plot ROPE with no Intercepts (cumulative model)

Open dontchaknowjo opened this issue 3 years ago • 4 comments

Question and context Hello, I was wondering if anyone has ideas about how to plot the ROPE analysis without plotting the intercepts.

here is what I tried that did not work:

percentage_in_rope1 <- rope(fit_sd, range=c(-0.1,0.1),ci=1)

percentage_in_rope1_2<- dplyr::filter(percentage_in_rope1, Parameter == "b_Hierarchy_ZeroSum")

plot(percentage_in_rope1_2)

dontchaknowjo avatar Jul 20 '22 16:07 dontchaknowjo

I also discovered that you can call show_intercept=FALSE, but this isn't affecting the plots of my rope analysis.

dontchaknowjo avatar Jul 21 '22 18:07 dontchaknowjo

@strengejacke Could we port the keep and drop arguments from parameters to the various bayestestR functions?

bwiernik avatar Sep 19 '22 13:09 bwiernik

@strengejacke Could we port the keep and drop arguments from parameters to the various bayestestR functions?

Yes, we could do that. However, show_intercept = FALSE should be available, there seems to be a bug in the plot() method. @dontchaknowjo Do you have a reproducible example?

strengejacke avatar Sep 19 '22 15:09 strengejacke

Yes thanks for asking. Let me know if you need anything else from me!

fit_interact_imp<- brm(important_manuscripts ~ Hierarchy_ZeroSum*self_MS_cooper+Identify+LogYears, data=TEST_data_S2, family=cumulative(probit), save_all_pars = TRUE, iter= 6000, warmup= 1000, thin= 1, inits = "0", prior=priors)

plot(rope(fit_interact),range=c(-0.1,0.1),ci=1,show_intercept=FALSE)+ ggplot2::labs(title = "Do you plan to share your next manuscript?")+xlim(-1,1)

On Sep 19, 2022, at 11:36 AM, Daniel @.***> wrote:

@strengejacke https://urldefense.com/v3/__https://github.com/strengejacke__;!!CzAuKJ42GuquVTTmVmPViYEvSg!KZAH-s5Wc4lrRUTuGeNfXMmSU4OyosZWlcQu8Z9NBlSVJS3UToRLY2Ik6kF-WmRReu27sSZNeJhyCApcmNEuBs0y$ Could we port the keep and drop arguments from parameters to the various bayestestR functions?

Yes, we could do that. However, show_intercept = FALSE should be available, there seems to be a bug in the plot() method. @dontchaknowjo https://urldefense.com/v3/__https://github.com/dontchaknowjo__;!!CzAuKJ42GuquVTTmVmPViYEvSg!KZAH-s5Wc4lrRUTuGeNfXMmSU4OyosZWlcQu8Z9NBlSVJS3UToRLY2Ik6kF-WmRReu27sSZNeJhyCApcmIbBaWsa$ Do you have a reproducible example?

— Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/easystats/bayestestR/issues/556*issuecomment-1251191231__;Iw!!CzAuKJ42GuquVTTmVmPViYEvSg!KZAH-s5Wc4lrRUTuGeNfXMmSU4OyosZWlcQu8Z9NBlSVJS3UToRLY2Ik6kF-WmRReu27sSZNeJhyCApcmBAybcVy$, or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AC4EGLIWUFYP6TGY2S5PXWLV7CCBPANCNFSM54EKP3TA__;!!CzAuKJ42GuquVTTmVmPViYEvSg!KZAH-s5Wc4lrRUTuGeNfXMmSU4OyosZWlcQu8Z9NBlSVJS3UToRLY2Ik6kF-WmRReu27sSZNeJhyCApcmKhCAkOv$. You are receiving this because you were mentioned.

dontchaknowjo avatar Oct 11 '22 07:10 dontchaknowjo