André Castro Bilbrough
@andreraoni.bsky.social
Data analyst who does evaluation for the NHS. I’m into Bayesianism (but exclusively in the bad vibes, cult kind of way), quasi-experimental methodology, pushing (just generally) and bad takes in all their forms.
Reposted by André Castro Bilbrough
I’m economically Bayesian but socially frequentist
October 9, 2025 at 10:50 PM
I’m economically Bayesian but socially frequentist
Reposted by André Castro Bilbrough
Ok so hear me out. Now that Gold Standard Science is becoming institutionalized with a second EO, here's our opportunity to set alpha to 1 for each and everything hypothesis test. Guaranteed 100% reproducibility, communicates uncertainty, falsifies every null hypothesis. Embodies the Gold Standard!
August 8, 2025 at 4:35 PM
Ok so hear me out. Now that Gold Standard Science is becoming institutionalized with a second EO, here's our opportunity to set alpha to 1 for each and everything hypothesis test. Guaranteed 100% reproducibility, communicates uncertainty, falsifies every null hypothesis. Embodies the Gold Standard!
Reposted by André Castro Bilbrough
This could fully pass for something on Clickhole, but it’s real, and I love it. Even down to his name.
July 15, 2025 at 1:17 PM
This could fully pass for something on Clickhole, but it’s real, and I love it. Even down to his name.
Reposted by André Castro Bilbrough
No one can stop you from rating the likert scale
November 6, 2024 at 7:57 PM
No one can stop you from rating the likert scale
Reposted by André Castro Bilbrough
Advanced analytics meets public value.
Our revolutionary NHP open-source demand model equips NHS analysts and planners with the tools to forecast demand, shift care to communities, and make evidence-driven decisions.
Built in the NHS, for the NHS.
www.strategyunitwm.nhs.uk/news/transfo...
Our revolutionary NHP open-source demand model equips NHS analysts and planners with the tools to forecast demand, shift care to communities, and make evidence-driven decisions.
Built in the NHS, for the NHS.
www.strategyunitwm.nhs.uk/news/transfo...
Transforming Hospital Planning with an Open-Source Demand and Capacity Model
We are proud to announce the open-sourcing of a demand and capacity model, developed with the New Hospitals Programme, to transform NHS hospital planning with transparency, collaboration, and effici...
www.strategyunitwm.nhs.uk
May 2, 2025 at 10:34 AM
Advanced analytics meets public value.
Our revolutionary NHP open-source demand model equips NHS analysts and planners with the tools to forecast demand, shift care to communities, and make evidence-driven decisions.
Built in the NHS, for the NHS.
www.strategyunitwm.nhs.uk/news/transfo...
Our revolutionary NHP open-source demand model equips NHS analysts and planners with the tools to forecast demand, shift care to communities, and make evidence-driven decisions.
Built in the NHS, for the NHS.
www.strategyunitwm.nhs.uk/news/transfo...
Reposted by André Castro Bilbrough
You. It’s gonna be you
the nice thing about creating a variable is that you never know who might be mad at you 20 years later for your weird decisions while doing so
April 21, 2025 at 5:44 PM
You. It’s gonna be you
Reposted by André Castro Bilbrough
💯The US fascist regime is wilfully destroying science. Europe must resist this and immunise itself against a tide of pseudoscience soon coming its way from the US.
March 27, 2025 at 3:00 PM
💯The US fascist regime is wilfully destroying science. Europe must resist this and immunise itself against a tide of pseudoscience soon coming its way from the US.
Reposted by André Castro Bilbrough
Check out this incredible stack exchange answer by the author of R package MatchIt on why/when matching is preferable to regression for causal inference:
stats.stackexchange.com/questions/54...
stats.stackexchange.com/questions/54...
Why do we do matching for causal inference vs regressing on confounders?
I'm new to the area of causal inference. From what I understand, one of the main concerns that causal inference tries to address is the effect of confounders!
For the sake of reference, let's denot...
stats.stackexchange.com
March 21, 2025 at 1:13 PM
Check out this incredible stack exchange answer by the author of R package MatchIt on why/when matching is preferable to regression for causal inference:
stats.stackexchange.com/questions/54...
stats.stackexchange.com/questions/54...
equally common is the experience of coming across something and saying, oh yea, my old friend, X, i know how this works, unexpected place to find this.
and then after digging, it turns out to be some completely unrelated thing, which just happens to be also called X for some reason
and then after digging, it turns out to be some completely unrelated thing, which just happens to be also called X for some reason
Me learning new stats things: “wow this sounds a lot like X”
Me after digging into the details: “This is just X but called Y now for some reason”
Me after digging into the details: “This is just X but called Y now for some reason”
March 4, 2025 at 8:41 PM
equally common is the experience of coming across something and saying, oh yea, my old friend, X, i know how this works, unexpected place to find this.
and then after digging, it turns out to be some completely unrelated thing, which just happens to be also called X for some reason
and then after digging, it turns out to be some completely unrelated thing, which just happens to be also called X for some reason
Reposted by André Castro Bilbrough
-- David Hume, A Treatise of Human Nature, 1739
If you're trying to gaslight me into believing that I'm not real, let me just say that it worked
no disrespect to my followers but i refuse to believe any of you are real. i say this with love.
January 14, 2025 at 5:37 PM
-- David Hume, A Treatise of Human Nature, 1739
Reposted by André Castro Bilbrough
Hangovers are the worst. More as we have it
January 1, 2025 at 8:11 AM
Hangovers are the worst. More as we have it
Reposted by André Castro Bilbrough
Reposted by André Castro Bilbrough
This is an insightful piece by Tim Bayne. It leads me to reflect on the contrast between Intentionality Dennett and Consciousness Dennett. To me, the ideas often seem like those of two different people... 🧵 (1/6) aeon.co/essays/as-re...
‘As real as it ever gets’: Dennett’s conception of the mind | Aeon Essays
Often metaphorical and allusive, the philosopher’s work will long be remembered for how it grappled with everyday thought
aeon.co
December 23, 2024 at 8:39 AM
This is an insightful piece by Tim Bayne. It leads me to reflect on the contrast between Intentionality Dennett and Consciousness Dennett. To me, the ideas often seem like those of two different people... 🧵 (1/6) aeon.co/essays/as-re...
Reposted by André Castro Bilbrough
Reposted by André Castro Bilbrough
Please consult your doctor if fuzzy estimands persist for more than 24 hours
October 16, 2024 at 3:38 PM
Please consult your doctor if fuzzy estimands persist for more than 24 hours
Reposted by André Castro Bilbrough
Reposted by André Castro Bilbrough
One thing I find depressing / unintentionally funny about the replication crisis in psychology is there’s all this high-level drama about retractions & fraud, then you look at the original study & it’s like… “Thinking about God made undergrads (N=28) wash their hands 0.8 seconds longer (p=0.04)”.
September 26, 2024 at 10:51 AM
One thing I find depressing / unintentionally funny about the replication crisis in psychology is there’s all this high-level drama about retractions & fraud, then you look at the original study & it’s like… “Thinking about God made undergrads (N=28) wash their hands 0.8 seconds longer (p=0.04)”.
Reposted by André Castro Bilbrough
I wish I was better at statistics.
September 10, 2024 at 4:29 PM
I wish I was better at statistics.
conference presentations on models.
frequentists: my research problem was x, so i built this model. this was useful because it allowed us to do y.
bayesian: this is my model, he’s a BITCH and i LOVE him. let me tell you the story of how i fixed all the STUPID SHIT he kept trying to do
frequentists: my research problem was x, so i built this model. this was useful because it allowed us to do y.
bayesian: this is my model, he’s a BITCH and i LOVE him. let me tell you the story of how i fixed all the STUPID SHIT he kept trying to do
July 24, 2024 at 8:21 AM
conference presentations on models.
frequentists: my research problem was x, so i built this model. this was useful because it allowed us to do y.
bayesian: this is my model, he’s a BITCH and i LOVE him. let me tell you the story of how i fixed all the STUPID SHIT he kept trying to do
frequentists: my research problem was x, so i built this model. this was useful because it allowed us to do y.
bayesian: this is my model, he’s a BITCH and i LOVE him. let me tell you the story of how i fixed all the STUPID SHIT he kept trying to do
Reposted by André Castro Bilbrough
Lifting to carry so much laundry it strikes fear into the hearts of all the neighbors.
July 8, 2024 at 8:01 PM
Lifting to carry so much laundry it strikes fear into the hearts of all the neighbors.
Reposted by André Castro Bilbrough
