Guido Biele
@guidobiele.bsky.social
Bayesian stats, causal inference, child and youth mental health.
https://gbiele.github.io/
https://gbiele.github.io/
wrote a bit fast, that's probably not an issue if one assumes the proportional odds assumtions holds and models the data accordingly, as done in the example.
If one relaxes the assumption and models category specific effects, things are probably different.
If one relaxes the assumption and models category specific effects, things are probably different.
August 25, 2025 at 3:34 PM
wrote a bit fast, that's probably not an issue if one assumes the proportional odds assumtions holds and models the data accordingly, as done in the example.
If one relaxes the assumption and models category specific effects, things are probably different.
If one relaxes the assumption and models category specific effects, things are probably different.
Very cool! I'm intrigued by the treatment of ordinal variables, and I have a question. Will this approach: avg_comparisons(mod_ord, variables = "partner", hypothesis = ~ I(sum(x * 1:5))) also give a good answer when the response-distribution is unbalanced, e.g. most people responded on level 1?
August 25, 2025 at 3:25 PM
Very cool! I'm intrigued by the treatment of ordinal variables, and I have a question. Will this approach: avg_comparisons(mod_ord, variables = "partner", hypothesis = ~ I(sum(x * 1:5))) also give a good answer when the response-distribution is unbalanced, e.g. most people responded on level 1?
If you are not discouraged by now, here is a relevant paper: Bayesian Identification and Estimation of Growth Mixture
Models (Xiao, Rabe-Heskeseth & Skrondal, 2025) www.duo.uio.no/bitstream/ha...
Models (Xiao, Rabe-Heskeseth & Skrondal, 2025) www.duo.uio.no/bitstream/ha...
Identifying Bayesian Mixture Models
betanalpha.github.io
August 13, 2025 at 10:54 AM
If you are not discouraged by now, here is a relevant paper: Bayesian Identification and Estimation of Growth Mixture
Models (Xiao, Rabe-Heskeseth & Skrondal, 2025) www.duo.uio.no/bitstream/ha...
Models (Xiao, Rabe-Heskeseth & Skrondal, 2025) www.duo.uio.no/bitstream/ha...
The question is well motivated, but there will be stumbling blocks or walls one can run into:
- degenerate classes / label switching (betanalpha.github.io/assets/case_...)
- finding # of latent classes
- non-informative class membership probs
I think very clean data are needed to have a chance.
- degenerate classes / label switching (betanalpha.github.io/assets/case_...)
- finding # of latent classes
- non-informative class membership probs
I think very clean data are needed to have a chance.
Identifying Bayesian Mixture Models
betanalpha.github.io
August 13, 2025 at 10:51 AM
The question is well motivated, but there will be stumbling blocks or walls one can run into:
- degenerate classes / label switching (betanalpha.github.io/assets/case_...)
- finding # of latent classes
- non-informative class membership probs
I think very clean data are needed to have a chance.
- degenerate classes / label switching (betanalpha.github.io/assets/case_...)
- finding # of latent classes
- non-informative class membership probs
I think very clean data are needed to have a chance.
No worries, das Deutsch ist nicht so gut, als dass man leicht Unterhaltungen auf Deutsch führen könnte 😀. (Bin weder verwandt noch verschwägert mit dem Eigentümer...)
August 10, 2025 at 1:39 PM
No worries, das Deutsch ist nicht so gut, als dass man leicht Unterhaltungen auf Deutsch führen könnte 😀. (Bin weder verwandt noch verschwägert mit dem Eigentümer...)
I recommend this places: maps.app.goo.gl/H2QkoDjDbJr8...
The owner speaks some German 😀. More importantly, it's a friendly and quiet place with reasonably priced very good meals and nice outdoor seating in the backyard.
The owner speaks some German 😀. More importantly, it's a friendly and quiet place with reasonably priced very good meals and nice outdoor seating in the backyard.
maps.app.goo.gl
August 10, 2025 at 9:46 AM
I recommend this places: maps.app.goo.gl/H2QkoDjDbJr8...
The owner speaks some German 😀. More importantly, it's a friendly and quiet place with reasonably priced very good meals and nice outdoor seating in the backyard.
The owner speaks some German 😀. More importantly, it's a friendly and quiet place with reasonably priced very good meals and nice outdoor seating in the backyard.
I think a "rich-get-richer" effect could be stronger if one does not provide additional info. But that would in my view also be bad practice. Asking an LLM to help with a review by using a RAG system with the relevant literature could well make the effect weaker.
August 1, 2025 at 2:10 PM
I think a "rich-get-richer" effect could be stronger if one does not provide additional info. But that would in my view also be bad practice. Asking an LLM to help with a review by using a RAG system with the relevant literature could well make the effect weaker.
Genau für diese Art von Text, der in erster Linie für Verwaltungszwecke verfasst wird, sind m. E. LLMs nützliche und effiziente Werkzeuge, bei denen, aus meiner Sicht, die Vorteile ggü den Nachteilen der Nutzung überwiegen.
July 28, 2025 at 9:01 AM
Genau für diese Art von Text, der in erster Linie für Verwaltungszwecke verfasst wird, sind m. E. LLMs nützliche und effiziente Werkzeuge, bei denen, aus meiner Sicht, die Vorteile ggü den Nachteilen der Nutzung überwiegen.
Generally agree. An neat use of ordinal regressions is to use them as a semi-parametric model for weird (multimodal) VAS data. If one splits the original scale in bins, one can with some effort calculate marginal effects on the original scale, with resolution depending on # of bins.
July 28, 2025 at 7:25 AM
Generally agree. An neat use of ordinal regressions is to use them as a semi-parametric model for weird (multimodal) VAS data. If one splits the original scale in bins, one can with some effort calculate marginal effects on the original scale, with resolution depending on # of bins.
Typos in the post are virtue signals, indicating that no LLM was used in its writing 😀
July 23, 2025 at 4:17 PM
Typos in the post are virtue signals, indicating that no LLM was used in its writing 😀
I don't see how the post effectively criticises target population estimands. It shows that if you have a target population estimand in mind, but your study isn't designed to estimate it, you'll pay with a high variance of your estimate.
July 14, 2025 at 4:00 PM
I don't see how the post effectively criticises target population estimands. It shows that if you have a target population estimand in mind, but your study isn't designed to estimate it, you'll pay with a high variance of your estimate.
OTH, “Our analysis should take into account that children grow around 7.5 cm per year and rarely less than 1 cm or more than 15 cm per year.” sounds entirely reasonable and illustrates that priors often reflect not just subjective beliefs but objective information that can robustify inference.
July 2, 2025 at 2:47 PM
OTH, “Our analysis should take into account that children grow around 7.5 cm per year and rarely less than 1 cm or more than 15 cm per year.” sounds entirely reasonable and illustrates that priors often reflect not just subjective beliefs but objective information that can robustify inference.
One reason for the lack of appreciation might be that Bayesian methods are often taught in a very abstract way, just as in the book at the start of the thread. Teaching the formula “posterior = prior × likelihood” tends to focus attention on so-called subjective beliefs as priors.
July 2, 2025 at 2:45 PM
One reason for the lack of appreciation might be that Bayesian methods are often taught in a very abstract way, just as in the book at the start of the thread. Teaching the formula “posterior = prior × likelihood” tends to focus attention on so-called subjective beliefs as priors.
I agree that one can use Bayes for many reasons and that it is useful in relatively simple situations. However, my (admittedly limited) observation is that the ability to fit complex models attracts more people to Bayesian estimation than improved inference in simpler cases.
July 2, 2025 at 12:49 PM
I agree that one can use Bayes for many reasons and that it is useful in relatively simple situations. However, my (admittedly limited) observation is that the ability to fit complex models attracts more people to Bayesian estimation than improved inference in simpler cases.
It works usually well if one uses t-tests or anovas to analyse simple experiments. But one shouldn't generalise from that to the goals/usefulness of Bayesian or Frequentist methods in general.
July 2, 2025 at 7:26 AM
It works usually well if one uses t-tests or anovas to analyse simple experiments. But one shouldn't generalise from that to the goals/usefulness of Bayesian or Frequentist methods in general.
It seems to be a common misunderstanding that the goal of applied Bayesian inference is to let the prior influence the inference. I think a more common goal is to be able to reliably fit complex models that are hard/impossible with Frequentist methods.
July 2, 2025 at 7:22 AM
It seems to be a common misunderstanding that the goal of applied Bayesian inference is to let the prior influence the inference. I think a more common goal is to be able to reliably fit complex models that are hard/impossible with Frequentist methods.
Did they look at both parental income and education? In Norwegian data the association of child outcomes with parental education is clearly stronger than with parental income.
June 26, 2025 at 3:43 PM
Did they look at both parental income and education? In Norwegian data the association of child outcomes with parental education is clearly stronger than with parental income.
To let LLMs produce less common and therefore likely also more creative text, one can always set the temperature parameter.
But there are trade offs. Less common might also be more likely wrong.
But there are trade offs. Less common might also be more likely wrong.
June 19, 2025 at 3:47 PM
To let LLMs produce less common and therefore likely also more creative text, one can always set the temperature parameter.
But there are trade offs. Less common might also be more likely wrong.
But there are trade offs. Less common might also be more likely wrong.
Sometimes psychologists collaborate with causal inference experts and use it for their research, here the probability of benefit or harm of ADHD medication for school outcomes: arxiv.org/abs/2502.10049
All math by @johandh2o.bsky.social
All math by @johandh2o.bsky.social
The Probability of Tiered Benefit: Partial Identification with Robust and Stable Inference
We define the Probability of Tiered Benefit in scenarios with a binary exposure and an outcome that is either categorical with $K \geq 2$ ordered tiers or continuous partitioned by $K-1$ fixed thresho...
share.google
June 18, 2025 at 3:24 PM
Sometimes psychologists collaborate with causal inference experts and use it for their research, here the probability of benefit or harm of ADHD medication for school outcomes: arxiv.org/abs/2502.10049
All math by @johandh2o.bsky.social
All math by @johandh2o.bsky.social
Understanding the transformer architecture helps explain why LLMs are so powerful—and when they’re likely to fail.
The Andrej Karpathy video goes into less detail about transformers than would be helpful. Here's a quick and accessible intro from 3Blue1Brown. www.youtube.com/watch?v=wjZo...
The Andrej Karpathy video goes into less detail about transformers than would be helpful. Here's a quick and accessible intro from 3Blue1Brown. www.youtube.com/watch?v=wjZo...
Transformers (how LLMs work) explained visually | DL5
YouTube video by 3Blue1Brown
www.youtube.com
June 18, 2025 at 10:53 AM
Understanding the transformer architecture helps explain why LLMs are so powerful—and when they’re likely to fail.
The Andrej Karpathy video goes into less detail about transformers than would be helpful. Here's a quick and accessible intro from 3Blue1Brown. www.youtube.com/watch?v=wjZo...
The Andrej Karpathy video goes into less detail about transformers than would be helpful. Here's a quick and accessible intro from 3Blue1Brown. www.youtube.com/watch?v=wjZo...
Easily only for models with identity link function (also careful with treatment-interactions). Otherwise one might have to code up a model in e.g. Stan, omit priors on coefficients involving the treatment & put a prior on the marginal effect (calculated in the transformed parameters block).
June 2, 2025 at 2:22 PM
Easily only for models with identity link function (also careful with treatment-interactions). Otherwise one might have to code up a model in e.g. Stan, omit priors on coefficients involving the treatment & put a prior on the marginal effect (calculated in the transformed parameters block).
The article talks about non-collapsibility, but doesn't seem to use the term. I think its a helpful term, because it puts a name to a key problem with OR, which is there even if the OR is not interpreted as a RR.
May 26, 2025 at 2:18 PM
The article talks about non-collapsibility, but doesn't seem to use the term. I think its a helpful term, because it puts a name to a key problem with OR, which is there even if the OR is not interpreted as a RR.