*nods* that's how you get multicolinearity
March 19, 2025 at 2:43 AM
*nods* that's how you get multicolinearity
Multicolinearity. What even is it?
July 27, 2026 at 1:57 PM
Multicolinearity. What even is it?
July 2, 2026 at 6:19 AM
“The GS3 consists of two closely related factors with high reliability.”
In a field that loves nuance and doesn’t know about multicolinearity, what could possibly go wrong …
In a field that loves nuance and doesn’t know about multicolinearity, what could possibly go wrong …
May 28, 2026 at 6:17 PM
“The GS3 consists of two closely related factors with high reliability.”
In a field that loves nuance and doesn’t know about multicolinearity, what could possibly go wrong …
In a field that loves nuance and doesn’t know about multicolinearity, what could possibly go wrong …
I missed this last week (my mentions turned into a dumpster fire) but yes, agree that this is a better way of thinking about it than just straight multicolinearity!
May 8, 2026 at 12:27 PM
I missed this last week (my mentions turned into a dumpster fire) but yes, agree that this is a better way of thinking about it than just straight multicolinearity!
*So* wrong. Many many years ago I encountered such a spectacularly wrong linear model in the wild that I still think about it. It illustrated almost every violation so perfectly (no the trite “multicolinearity blows up standard errors” but fundamentally fancy means are still fragile means)
July 9, 2026 at 1:45 AM
*So* wrong. Many many years ago I encountered such a spectacularly wrong linear model in the wild that I still think about it. It illustrated almost every violation so perfectly (no the trite “multicolinearity blows up standard errors” but fundamentally fancy means are still fragile means)
#statstab #613 Additional caution regarding rules of thumb for variance inflation factors: extending O’Brien to the context of specification error
Thoughts: I'm sure some will have strong opinions on VIF.
#VIF #regression #ruleofthumb #multicolinearity
link.springer.com/article/10.1...
Thoughts: I'm sure some will have strong opinions on VIF.
#VIF #regression #ruleofthumb #multicolinearity
link.springer.com/article/10.1...
Additional caution regarding rules of thumb for variance inflation factors: extending O’Brien to the context of specification error - Quality & Quantity
Despite the popularity of Variance Inflation Factor (VIF) scores as multicollinearity diagnostics, methodological work has voiced skepticism. In a paper that has become the most highly cited reference...
link.springer.com
September 9, 2026 at 7:09 PM
#statstab #613 Additional caution regarding rules of thumb for variance inflation factors: extending O’Brien to the context of specification error
Thoughts: I'm sure some will have strong opinions on VIF.
#VIF #regression #ruleofthumb #multicolinearity
link.springer.com/article/10.1...
Thoughts: I'm sure some will have strong opinions on VIF.
#VIF #regression #ruleofthumb #multicolinearity
link.springer.com/article/10.1...
#statstab #506 The VIF Score. What is it Good For? Absolutely Nothing
Thoughts: Maybe we need to think more closely about what these metrics mean.
#diagnostics #assumptions #regression #vif #variance #variableselection #error #multicolinearity
journals.sagepub.com/doi/abs/10.1...
Thoughts: Maybe we need to think more closely about what these metrics mean.
#diagnostics #assumptions #regression #vif #variance #variableselection #error #multicolinearity
journals.sagepub.com/doi/abs/10.1...
The VIF Score. What is it Good For? Absolutely Nothing - Arturs Kalnins, Kendall Praitis Hill, 2025
Variance inflation factors (VIF scores) are regression diagnostics commonly invoked throughout the social sciences. Researchers typically take the perspective t...
journals.sagepub.com
March 16, 2026 at 8:58 PM
#statstab #506 The VIF Score. What is it Good For? Absolutely Nothing
Thoughts: Maybe we need to think more closely about what these metrics mean.
#diagnostics #assumptions #regression #vif #variance #variableselection #error #multicolinearity
journals.sagepub.com/doi/abs/10.1...
Thoughts: Maybe we need to think more closely about what these metrics mean.
#diagnostics #assumptions #regression #vif #variance #variableselection #error #multicolinearity
journals.sagepub.com/doi/abs/10.1...
These are among my least favorite jobs, because they arise from no clear research question but having tons of pre-existing data.
Most of the job is pruning out variables due to cell size and multicolinearity.
I think my resolve to keep stats consulting dies a little bit more each time I do one.
Most of the job is pruning out variables due to cell size and multicolinearity.
I think my resolve to keep stats consulting dies a little bit more each time I do one.
February 4, 2024 at 12:55 PM
These are among my least favorite jobs, because they arise from no clear research question but having tons of pre-existing data.
Most of the job is pruning out variables due to cell size and multicolinearity.
I think my resolve to keep stats consulting dies a little bit more each time I do one.
Most of the job is pruning out variables due to cell size and multicolinearity.
I think my resolve to keep stats consulting dies a little bit more each time I do one.
I think this works as an intuitive explanation for multicolinearity, right? When the cloud is flat the plane is attached in a very stable way. The table has many legs. But when we flatten that cloud with an association between X2 and X2 the table becomes "wobbly" orthogonal to that axis.
October 14, 2025 at 9:07 AM
I think this works as an intuitive explanation for multicolinearity, right? When the cloud is flat the plane is attached in a very stable way. The table has many legs. But when we flatten that cloud with an association between X2 and X2 the table becomes "wobbly" orthogonal to that axis.
So I thought of a visualization on the train for how multicolinearity works (maybe) but now I don’t have any computer access do make it/check it and I have clinical work all day. Pain.
October 10, 2025 at 6:18 AM
So I thought of a visualization on the train for how multicolinearity works (maybe) but now I don’t have any computer access do make it/check it and I have clinical work all day. Pain.
F*** multicolinearity
November 16, 2024 at 7:55 PM
F*** multicolinearity
Like, this has to be because of the wealth effects, right? A case where controlling for wealth would be too obvious multicolinearity?
@caseybreen.bsky.social, in “Longevity Benefits of Homeownership,” finds life expectancy differs btw homeowners & renters, presenting “compelling evidence that owning a home in early adulthood has a causal effect on later-life mortality.” @oxforddemsci.bsky.social read.dukeupress.edu/demography/a...
December 5, 2024 at 6:46 PM
Like, this has to be because of the wealth effects, right? A case where controlling for wealth would be too obvious multicolinearity?
Thank you for the additional analysis. Multicolinearity is a thorn.
February 25, 2024 at 3:46 PM
Thank you for the additional analysis. Multicolinearity is a thorn.
Another way to look and consider multicollinearity. Check your VIF.
Linear Regression: Multicolinearity, they taught you wrong by Yukio https://link.medium.com/NAGugZ0aIEb
Linear Regression: Multicolinearity, they taught you wrong by Yukio https://link.medium.com/NAGugZ0aIEb
November 22, 2024 at 12:41 PM
Another way to look and consider multicollinearity. Check your VIF.
Linear Regression: Multicolinearity, they taught you wrong by Yukio https://link.medium.com/NAGugZ0aIEb
Linear Regression: Multicolinearity, they taught you wrong by Yukio https://link.medium.com/NAGugZ0aIEb
Ooh weird q but is there any multicolinearity there? E.g. moderates fundraise better or worse?
August 21, 2025 at 11:38 PM
Ooh weird q but is there any multicolinearity there? E.g. moderates fundraise better or worse?
And where might be put multicolinearity on this figure? 🛋️🍿
August 18, 2025 at 9:05 PM
And where might be put multicolinearity on this figure? 🛋️🍿
... the intuition works here.
3. As usual, I believe the fear of multicolinearity --at least in business analysis contexts-- is driven by management expectations: managers want causal levers without understanding of wanting to invest what it takes to look for them... 3/
3. As usual, I believe the fear of multicolinearity --at least in business analysis contexts-- is driven by management expectations: managers want causal levers without understanding of wanting to invest what it takes to look for them... 3/
October 13, 2025 at 11:12 PM
... the intuition works here.
3. As usual, I believe the fear of multicolinearity --at least in business analysis contexts-- is driven by management expectations: managers want causal levers without understanding of wanting to invest what it takes to look for them... 3/
3. As usual, I believe the fear of multicolinearity --at least in business analysis contexts-- is driven by management expectations: managers want causal levers without understanding of wanting to invest what it takes to look for them... 3/
That seems like a minefield of multicolinearity problems. 🙃
September 9, 2024 at 10:05 PM
That seems like a minefield of multicolinearity problems. 🙃
Anybody do a regression analysis with SDoH as covariates? Lots of multicolinearity with these vars... any luck with ridge regression?
October 7, 2023 at 7:59 PM
Anybody do a regression analysis with SDoH as covariates? Lots of multicolinearity with these vars... any luck with ridge regression?