brms Version: 2.10.0 Hi, This is my first foray into non-normal modelling, so please bear with me if I’m asking obvious questions. I have searched the forum and looked at various documentations, but I remain unclear on a couple of things. I have data which can be considered long reaction times (…

Hey! Looking at the actual Stan code generated by make_stancode shows that the shifted Log-Normal is parameterized like lognormal_lpdf(Y - ndt | mu, sigma), so, ignoring Jacobians for a moment, this means \log(Y - \text{ndt}) \sim \text{Normal}(\mu, \sigma). With link identity for mu, the coefficie…

Hey Anthony! Welcome to the Stan forum! Sorry it took us so long to respond. Disclaimer: I rarely work with brms, and never used the shifted log-normal (is ndt the “shift-model”?). Generally, the estimates of the Log-Normal are on the log-scale. [image] AnthonyGabay: Paul seems to suggest that…

Hi Max, Thanks for your reply! I hope you don’t mind if I reply with some follow up questions/comments. Yes you are right, ndt is the shift parameter. Your message prompted me to look deeper into the brms families documentation and code , and this is what I found: shifted_lognormal <- function(lin…

Hi Max, This makes so much sense. Thanks for the clear explanation, I really appreciate it. Cheers, Anthony

Hi, I have read this thread and the thread that is linked to this one about the shifted lognormal model for which I am also doing a prior predictive check. However, after reading both threads I still don’t understand how the model works. I have now simplified my model and excluded random effects t…

Hey @Max_Mantei Are you sure that truncation works in distributional models? I have a similar model to this, and based on my prior predictive distributions, truncation does not work for distributional models.

Hi @Ross_Kempner ! Sorry for the delayed response. I think truncation should work also for distributional models. I’m not sure if prior predictive checks are implemented in BRMS, but I’m pretty sure it’d be possible in Stan. Check out the the relevant section in the user manual . Cheers, Max

Sorry that I didn’t see you post earlier. I’m actually not sure how to answer. I think it might be helpful (albeit tedious) to look at the generated Stan code and try to reverse engineer it in R. Another point you might have to consider is that looking at point estimates/summaries of expectations c…

Understanding shifted lognormal parameters and priors

Interfaces brms

Topic		Replies	Views
Shifted log normal distribution with varying shifts by participant brms	12	3475	November 23, 2020
Fitting brm shifted lognormal model for reaction times Modeling fitting-issues , brms	3	163	January 29, 2025
Weakly informative priors for Shifted LogNormal Modeling specification , prior-choice , cognitive-science	3	1105	March 1, 2021
Advice on distribution for modelling reaction times? (and setting priors) brms cognitive-science	8	1911	November 23, 2020
Shifted lognormal stan conversion and mean function brms	4	593	December 5, 2020

Understanding shifted lognormal parameters and priors

Related topics