Discussion
Loading...

#Tag

Log in
  • About
  • Code of conduct
  • Privacy
  • Users
  • Instances
  • About Bonfire
Fabrizio Musacchio
Fabrizio Musacchio
@FabMusacchio@mastodon.social  ·  activity timestamp last week

Due to a recent discussion with colleagues on whether and when to use #LinearMixedModels ( #LMM), I wrote a blog post comparing LMM to other approaches using simulated data. I thought, it may also be useful for others working with hierarchical data structures in #neuroscience and beyond.

🌍 https://www.fabriziomusacchio.com/blog/2026-01-31-linear_mixed_models/

#Python #Statistics #DataScience #MixedModels #Statsmodels #ANOVA #ANCOVA #GLMM #regression

2 media
Line plot of simulated neural responses versus stimulus strength for three subjects, with residuals and group comparisons.
Line plot of simulated neural responses versus stimulus strength for three subjects, with residuals and group comparisons.
Line plot of simulated neural responses versus stimulus strength for three subjects, with residuals and group comparisons.
Group specific slope comparison in an LMM friendly regime with many groups and few observations per group. ANCOVA interaction slope estimates scatter widely, while LMM BLUP slopes are pulled toward the population mean.
Group specific slope comparison in an LMM friendly regime with many groups and few observations per group. ANCOVA interaction slope estimates scatter widely, while LMM BLUP slopes are pulled toward the population mean.
Group specific slope comparison in an LMM friendly regime with many groups and few observations per group. ANCOVA interaction slope estimates scatter widely, while LMM BLUP slopes are pulled toward the population mean.
Fabrizio Musacchio

Linear mixed models in practice: When ANCOVA is enough and when you really need random effects

Linear mixed models (LMMs) are a powerful statistical tool for analyzing hierarchical or grouped data, common in neuroscience experiments. This post provides a practical guide on when to use LMMs versus traditional ANCOVA approaches, highlighting the advantages of mixed models in handling dependencies, unbalanced designs, and stabilizing estimates through shrinkage. Through simulated examples, we illustrate the differences in model performance and interpretation, helping you to make informed decisions about your statistical analyses.
  • Copy link
  • Flag this post
  • Block
Em :official_verified: boosted
Malka Beth
Malka Beth
@BrahmaBelarusian@regenerate.social  ·  activity timestamp 3 weeks ago

@Em0nM4stodon same, I really feel the approximate peak of modern technology in it's utility, ease of use, valued cost & lack of privacy invasions, was reached around 2000-2005 ACE. By then we had solid speech/talk to text software & hardware. Mobile phones, including smartphones were durable & had cameras or not depending on your preference. Similarly phones had GPS, & satellite or not. Screen readers were likewise fully developed.

Not everyone needs everything a piece of advanced technology & that's why I would love to see a renewed separation of different devices. Cheap mapping/direction GPS systems, separate from phones, separate from PDAs, separate from cameras.

Speed of doing every tech related task hasn't really changed for most people & if anything it's probably slower due to infrastructure decay.

*As I'm unfortunately continuing to consider which smartphone I should get over the next year & same for my spouse, it really hits me hard how it'd be better for us both if we could just get PDAs, his with a touchscreen & me with a physical keyboard that were 5g enabled.
#Tech #NoGenAI #SurveillanceState #Innovation #Regression #DisabilityAccessibility

  • Copy link
  • Flag this post
  • Block
Em :official_verified:
Em :official_verified:
@Em0nM4stodon@infosec.exchange  ·  activity timestamp 3 weeks ago

I used to be excited about new tech and new gadgets. Now I'm just scared it's going to be more surveillance and more AI shenanigans.

long sigh

#Tech #NoAI

Malka Beth
Malka Beth
@BrahmaBelarusian@regenerate.social replied  ·  activity timestamp 3 weeks ago

@Em0nM4stodon same, I really feel the approximate peak of modern technology in it's utility, ease of use, valued cost & lack of privacy invasions, was reached around 2000-2005 ACE. By then we had solid speech/talk to text software & hardware. Mobile phones, including smartphones were durable & had cameras or not depending on your preference. Similarly phones had GPS, & satellite or not. Screen readers were likewise fully developed.

Not everyone needs everything a piece of advanced technology & that's why I would love to see a renewed separation of different devices. Cheap mapping/direction GPS systems, separate from phones, separate from PDAs, separate from cameras.

Speed of doing every tech related task hasn't really changed for most people & if anything it's probably slower due to infrastructure decay.

*As I'm unfortunately continuing to consider which smartphone I should get over the next year & same for my spouse, it really hits me hard how it'd be better for us both if we could just get PDAs, his with a touchscreen & me with a physical keyboard that were 5g enabled.
#Tech #NoGenAI #SurveillanceState #Innovation #Regression #DisabilityAccessibility

  • Copy link
  • Flag this comment
  • Block
Natanael Copa
Natanael Copa
@ncopa@fosstodon.org  ·  activity timestamp 5 months ago

OpenSSL 3.5.3 introduces a regression that makes ssh break on openssl version check.
https://github.com/openssl/project/issues/1621

If you use Alpine Linux:
- 3.22: make sure you get >=3.5.3-r1
- edge: make sure you get >=3.5.3-r2

Those include backports of upstream fixes.

Sorry for the inconvenience.

#AlpineLinux #OpenSSL #regression

  • Copy link
  • Flag this post
  • Block
Pierre Lebeaupin
Pierre Lebeaupin
@PierreLebeaupin@mastodon.gougere.fr  ·  activity timestamp 7 months ago

Non, c’est clairement la #régression au moins perçue, et le « tombage de haut » qui l’accompagne, qui expliquent beaucoup plus aisément la #mobilisation.

Mais si une mobilisation exceptionnelle faisait qu’une politique était souhaitable à appliquer, ou même majoritaire, ça se saurait.

Les gilets jaunes aussi pouvaient se prévaloir d’une mobilisation exceptionnelle. Est-ce qu’il était souhaitable d’appliquer leur projet politique pour autant? ⤵️

  • Copy link
  • Flag this post
  • Block
🫧 socialcoding.. boosted
Kevin Karhan :verified:
Kevin Karhan :verified:
@kkarhan@infosec.space  ·  activity timestamp 7 months ago

One thing that really pisses me off personally is the #regression in terms of #Messenger#Apps.

My personal distaste and dislike for #proprietary, #SingleVendor & #SingleProvider #services like #Signal¹, #Telegram, #Discord², #WhatsApp, #Slack, #MicrosoftTeams, etc. aside:

  • WHY is there no #CrossProvider#Messenger to handle that shite?

  • WHY does everyone of these shitty providers think people want to download their #bloated#WebApp that takes up triple digit Megabytes if not entire Gigabytes and will gobble up all the #RAM and #CPU they can??

This problem ain't new and already got solved for corporate social media ages ago! (Not to mention actually good messengers!)

  • And no, bridgesdon't count!

  • I mean API 0 - style access because obviously none of the platforms will allow, endorse or support such an endeavour and actively fight the developers and users !

So yeah, consider this a call for a @gajim / #Gajim or @pidgin / #Pidginfor garbage platforms!

  • Cuz back in the day we had way worse messengers yet people actually made #AIM, #ICQ, #MSN, #QQ, #IRC & #XMPP work just fine from one single "phat" client!

  • Can we please get that back?

#api0#Enshittification

  • Copy link
  • Flag this post
  • Block
Kevin Karhan :verified:
Kevin Karhan :verified:
@kkarhan@infosec.space  ·  activity timestamp 7 months ago

One thing that really pisses me off personally is the #regression in terms of #Messenger#Apps.

My personal distaste and dislike for #proprietary, #SingleVendor & #SingleProvider #services like #Signal¹, #Telegram, #Discord², #WhatsApp, #Slack, #MicrosoftTeams, etc. aside:

  • WHY is there no #CrossProvider#Messenger to handle that shite?

  • WHY does everyone of these shitty providers think people want to download their #bloated#WebApp that takes up triple digit Megabytes if not entire Gigabytes and will gobble up all the #RAM and #CPU they can??

This problem ain't new and already got solved for corporate social media ages ago! (Not to mention actually good messengers!)

  • And no, bridgesdon't count!

  • I mean API 0 - style access because obviously none of the platforms will allow, endorse or support such an endeavour and actively fight the developers and users !

So yeah, consider this a call for a @gajim / #Gajim or @pidgin / #Pidginfor garbage platforms!

  • Cuz back in the day we had way worse messengers yet people actually made #AIM, #ICQ, #MSN, #QQ, #IRC & #XMPP work just fine from one single "phat" client!

  • Can we please get that back?

#api0#Enshittification

  • Copy link
  • Flag this post
  • Block
Aral Balkan
Aral Balkan
@aral@mastodon.ar.al  ·  activity timestamp 7 months ago

🥳 New Kitten Release

• Fixed regression: Since we switched the h() render function to return a generator (for seamless async support in html templates), two other methods that were calling h() directly – kitten.safelyAddHtml() and kitten.js() – had begun to fail. This was also affecting the fetchiverse, streamiverse, and kitten-chat family of examples.

Enjoy!

https://kitten.small-web.org

(Run kitten update to immediately install the latest version on your development machines. Your production machines will automatically update in the next few hours.)

kitten💕

#Kitten #SmallWeb #bug #regression #fix#KittenRelease#SmallTech #web #dev

  • Copy link
  • Flag this post
  • Block

bonfire.cafe

A space for Bonfire maintainers and contributors to communicate

bonfire.cafe: About · Code of conduct · Privacy · Users · Instances
Bonfire social · 1.0.2-alpha.22 no JS en
Automatic federation enabled
Log in
  • Explore
  • About
  • Members
  • Code of Conduct