@bggj.bsky.social
PhD student of statistics at University of Iceland focusing on statistics of extremes, spatial statistics, and copulas, but interested in most of statistics.
Personal website: https://bggj.is
Personal website: https://bggj.is
Reposted
What is green, angry, and can be computed in near-linear time?
The convex Hulk
The convex Hulk
January 11, 2025 at 8:12 PM
What is green, angry, and can be computed in near-linear time?
The convex Hulk
The convex Hulk
Reposted
Finally compiled and refined this a bit; do take a look if you're curious! A more TeX'd version is attached in the screenshots.
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
December 30, 2024 at 3:42 PM
Finally compiled and refined this a bit; do take a look if you're curious! A more TeX'd version is attached in the screenshots.
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
New post on using copulas in Stan! This one is about the t-copula.
bggj.is/posts/t-copu...
#stats #rstats
bggj.is/posts/t-copu...
#stats #rstats
It was the best of tails, it was the worst of tails: The T-Copula – bggj
The Gaussian copula’s inability to model tail dependence can be a serious limitation in practice. This post introduces the t-copula, which shares many convenient properties with the Gaussian copula wh...
bggj.is
December 12, 2024 at 1:21 PM
New post on using copulas in Stan! This one is about the t-copula.
bggj.is/posts/t-copu...
#stats #rstats
bggj.is/posts/t-copu...
#stats #rstats
Just a heads up: If you're not using version control, never ever ever write "output-dir: ." or similar in your _quarto.yml file. It will delete everything in the base folder recursively (since it wants to refresh the output with this new setting).
Luckily, I was using version control.
Luckily, I was using version control.
October 25, 2024 at 7:56 AM
Just a heads up: If you're not using version control, never ever ever write "output-dir: ." or similar in your _quarto.yml file. It will delete everything in the base folder recursively (since it wants to refresh the output with this new setting).
Luckily, I was using version control.
Luckily, I was using version control.
Just a little more than a month until parliamentary elections here in Iceland.
I maintain an easily updated Quarto data-website in Icelandic.
metill.is
The repo is here: github.com/metill-is/me...
Main takeaways: #quarto + EJS to automatically update front page with new content.
#rstats
I maintain an easily updated Quarto data-website in Icelandic.
metill.is
The repo is here: github.com/metill-is/me...
Main takeaways: #quarto + EJS to automatically update front page with new content.
#rstats
October 22, 2024 at 1:55 PM
Just a little more than a month until parliamentary elections here in Iceland.
I maintain an easily updated Quarto data-website in Icelandic.
metill.is
The repo is here: github.com/metill-is/me...
Main takeaways: #quarto + EJS to automatically update front page with new content.
#rstats
I maintain an easily updated Quarto data-website in Icelandic.
metill.is
The repo is here: github.com/metill-is/me...
Main takeaways: #quarto + EJS to automatically update front page with new content.
#rstats
Reposted
The latest `brms` CRAN release added support for `priorsense` for easy prior and likelihood sensitivity analysis doi.org/10.1007/s112...
```
> fit |>
powerscale_plot_dens(variable='b_doseg', help_text=FALSE) +
labs(x='Dose (g) coefficient', y=NULL)
```
#Bayesian #rstats
```
> fit |>
powerscale_plot_dens(variable='b_doseg', help_text=FALSE) +
labs(x='Dose (g) coefficient', y=NULL)
```
#Bayesian #rstats
October 3, 2024 at 8:07 AM
The latest `brms` CRAN release added support for `priorsense` for easy prior and likelihood sensitivity analysis doi.org/10.1007/s112...
```
> fit |>
powerscale_plot_dens(variable='b_doseg', help_text=FALSE) +
labs(x='Dose (g) coefficient', y=NULL)
```
#Bayesian #rstats
```
> fit |>
powerscale_plot_dens(variable='b_doseg', help_text=FALSE) +
labs(x='Dose (g) coefficient', y=NULL)
```
#Bayesian #rstats
Reposted
September 24, 2024 at 8:30 AM
The second post in my series on copulas in Stan is up.
In this one I define the Gaussian copula, show how to use it in Stan with a small example, then list some of its positives and negatives.
bggj.is/posts/gaussi...
In this one I define the Gaussian copula, show how to use it in Stan with a small example, then list some of its positives and negatives.
bggj.is/posts/gaussi...
A Gentle Introduction: The Gaussian Copula – bggj
This post continues the series on copulas in Stan by introducing the Gaussian copula, discussing its properties, applications, and providing examples of how to implement it in Stan.
bggj.is
September 23, 2024 at 2:16 PM
The second post in my series on copulas in Stan is up.
In this one I define the Gaussian copula, show how to use it in Stan with a small example, then list some of its positives and negatives.
bggj.is/posts/gaussi...
In this one I define the Gaussian copula, show how to use it in Stan with a small example, then list some of its positives and negatives.
bggj.is/posts/gaussi...
Generative neutral network subscription site called OnlyGAN
Spline modelling subscription site called OnlyGAMs
Bayesian modeling subscription site called OnlyStan
September 22, 2024 at 9:36 PM
Generative neutral network subscription site called OnlyGAN
Reposted
Can you solve these three vaccine puzzles?
New post on the counterintuitive features of vaccine rollouts during dynamic epidemics: blog.kucharski.io?three-vaccin...
New post on the counterintuitive features of vaccine rollouts during dynamic epidemics: blog.kucharski.io?three-vaccin...
September 20, 2024 at 6:57 AM
Can you solve these three vaccine puzzles?
New post on the counterintuitive features of vaccine rollouts during dynamic epidemics: blog.kucharski.io?three-vaccin...
New post on the counterintuitive features of vaccine rollouts during dynamic epidemics: blog.kucharski.io?three-vaccin...
Reposted
I'm looking for a post-doc and doctoral student to join my group at Aalto to work on Bayesian workflow, cross-validation, model checking, projection predictive model selection, inference diagnostics, priors, and survival analysis (I have plenty of research ideas, pick any combination)
#Bayesian
#Bayesian
September 20, 2024 at 5:50 AM
I'm looking for a post-doc and doctoral student to join my group at Aalto to work on Bayesian workflow, cross-validation, model checking, projection predictive model selection, inference diagnostics, priors, and survival analysis (I have plenty of research ideas, pick any combination)
#Bayesian
#Bayesian
Reposted
Check out @stephenjwild.bsky.social's Bayesian starter pack go.bsky.app/UfgChHe
September 18, 2024 at 1:25 AM
Check out @stephenjwild.bsky.social's Bayesian starter pack go.bsky.app/UfgChHe
I was so inspired after my first RSS Conference two weeks ago that I hurried up and became a fellow when I came back home to Iceland this week!
Happy to join this large community of amazing statisticians!
Happy to join this large community of amazing statisticians!
September 18, 2024 at 12:02 PM
I was so inspired after my first RSS Conference two weeks ago that I hurried up and became a fellow when I came back home to Iceland this week!
Happy to join this large community of amazing statisticians!
Happy to join this large community of amazing statisticians!
After a great #StanCon2024, I've started a blog series on using copulas in Stan.
This first post gives a quick introduction to what copulas are. I'll keep each post short but over time I hope to put together a repository of helpful content!
bggj.is/posts/stan-c...
#stats
This first post gives a quick introduction to what copulas are. I'll keep each post short but over time I hope to put together a repository of helpful content!
bggj.is/posts/stan-c...
#stats
If It Bleeds, We Can Kill It – bggj
This is the first post in what’s going to be a series on using Copulas in Stan. Each post is going to be short to keep me from postponing writing them. In this post I lightly introduce the series and ...
bggj.is
September 17, 2024 at 11:24 AM
After a great #StanCon2024, I've started a blog series on using copulas in Stan.
This first post gives a quick introduction to what copulas are. I'll keep each post short but over time I hope to put together a repository of helpful content!
bggj.is/posts/stan-c...
#stats
This first post gives a quick introduction to what copulas are. I'll keep each post short but over time I hope to put together a repository of helpful content!
bggj.is/posts/stan-c...
#stats
Reposted
To be fair people of all generations struggle with properly interpreting and manipulating probability density functions.
August 28, 2024 at 4:02 PM
To be fair people of all generations struggle with properly interpreting and manipulating probability density functions.
Looking forward to talking about Gaussian Markov Random Fields and Copulas at StanCon '24 at Oxford University in September!
Early bird registration is still live, so get your tickets while they're cheaper: mc-stan.org/events/stanc...
Early bird registration is still live, so get your tickets while they're cheaper: mc-stan.org/events/stanc...
July 4, 2024 at 8:16 PM
Looking forward to talking about Gaussian Markov Random Fields and Copulas at StanCon '24 at Oxford University in September!
Early bird registration is still live, so get your tickets while they're cheaper: mc-stan.org/events/stanc...
Early bird registration is still live, so get your tickets while they're cheaper: mc-stan.org/events/stanc...
Reposted
StanCon is still happening in September. Have you considered the probabilistic dividends? The keynote speakers will cover everything from disease to drugs to sharks. Bayesian Sharks (doo-doo, doo-doo, doo-doo). #stats Details: t.co/EuEqamK2UR
June 18, 2024 at 2:15 PM
StanCon is still happening in September. Have you considered the probabilistic dividends? The keynote speakers will cover everything from disease to drugs to sharks. Bayesian Sharks (doo-doo, doo-doo, doo-doo). #stats Details: t.co/EuEqamK2UR
Reposted
Happy Icelandic independence day! Here is the Icelandic flag in polar coordinates because it is the nerdiest thing I have seen this month. Credit goes to my former student Binni. #17júní @bggj.bsky.social
June 17, 2024 at 3:02 PM
Happy Icelandic independence day! Here is the Icelandic flag in polar coordinates because it is the nerdiest thing I have seen this month. Credit goes to my former student Binni. #17júní @bggj.bsky.social
Siggi er í alveg trylltum herringbone jakka sem virðist vera þægilegur og sjúklega vel sniðinn. En Siggi, vinsamlegast hnepptu frá tölunni! #vikan
December 1, 2023 at 9:51 PM
Siggi er í alveg trylltum herringbone jakka sem virðist vera þægilegur og sjúklega vel sniðinn. En Siggi, vinsamlegast hnepptu frá tölunni! #vikan
Vá hvað 'Sucks to be you, Nigel' voru geggjuð í Vikunni. Þetta hljómaði eins og eðal 90s alternative rock í anda Slint og June of 44.
Strax búinn að kaupa plötuna þeirra: spectralassaultrecords.bandcamp.com/album/t-na-b...
Strax búinn að kaupa plötuna þeirra: spectralassaultrecords.bandcamp.com/album/t-na-b...
Tína blóm, by Sucks to be you, Nigel
10 track album
spectralassaultrecords.bandcamp.com
November 10, 2023 at 9:47 PM
Vá hvað 'Sucks to be you, Nigel' voru geggjuð í Vikunni. Þetta hljómaði eins og eðal 90s alternative rock í anda Slint og June of 44.
Strax búinn að kaupa plötuna þeirra: spectralassaultrecords.bandcamp.com/album/t-na-b...
Strax búinn að kaupa plötuna þeirra: spectralassaultrecords.bandcamp.com/album/t-na-b...
Reposted
Have you ever found #rstats rowwise computations cumbersome, slow, or memory greedy?
If so, do give a try to **lay** -- a small 📦 from @romainfrancois.bsky.social and myself.
Check courtiol.github.io/lay/ for examples, alternatives and benchmarks.
You can use **lay** alone, or with **dplyr** 👀👇
If so, do give a try to **lay** -- a small 📦 from @romainfrancois.bsky.social and myself.
Check courtiol.github.io/lay/ for examples, alternatives and benchmarks.
You can use **lay** alone, or with **dplyr** 👀👇
October 31, 2023 at 9:09 AM
Have you ever found #rstats rowwise computations cumbersome, slow, or memory greedy?
If so, do give a try to **lay** -- a small 📦 from @romainfrancois.bsky.social and myself.
Check courtiol.github.io/lay/ for examples, alternatives and benchmarks.
You can use **lay** alone, or with **dplyr** 👀👇
If so, do give a try to **lay** -- a small 📦 from @romainfrancois.bsky.social and myself.
Check courtiol.github.io/lay/ for examples, alternatives and benchmarks.
You can use **lay** alone, or with **dplyr** 👀👇
Ég er að bæta greiningu á "breiðum grunni verðbólgu" í sjálfvirku skýrsluna um vísitölu neysluverðs á metill.is.
Er þetta nógu skiljanleg framsetning? Er auðvelt að meðtaka þessar upplýsingar?
Er þetta nógu skiljanleg framsetning? Er auðvelt að meðtaka þessar upplýsingar?
October 30, 2023 at 1:50 PM
Ég er að bæta greiningu á "breiðum grunni verðbólgu" í sjálfvirku skýrsluna um vísitölu neysluverðs á metill.is.
Er þetta nógu skiljanleg framsetning? Er auðvelt að meðtaka þessar upplýsingar?
Er þetta nógu skiljanleg framsetning? Er auðvelt að meðtaka þessar upplýsingar?
Í nýjustu útgáfunni af Quarto er núna mega auðvelt að setja saman dashboards. Bless bless frítími.
quarto.org/docs/dashboa...
quarto.org/docs/dashboa...
October 25, 2023 at 7:29 PM
Í nýjustu útgáfunni af Quarto er núna mega auðvelt að setja saman dashboards. Bless bless frítími.
quarto.org/docs/dashboa...
quarto.org/docs/dashboa...
Reposted
Áfram stelpur og stálp! Gerum þjóðfélagið betra!
October 24, 2023 at 2:15 PM
Áfram stelpur og stálp! Gerum þjóðfélagið betra!