Moritz Schauer
@mschauer.bsky.social
Statistician, Associate Professor (Lektor) at University of Gothenburg and Chalmers; inference and conditional distributions for anything
https://mschauer.github.io
http://orcid.org/0000-0003-3310-7915
[ˈmoː/r/ɪts ˈʃaʊ̯ɐ]
https://mschauer.github.io
http://orcid.org/0000-0003-3310-7915
[ˈmoː/r/ɪts ˈʃaʊ̯ɐ]
The event B having happened is not the same as the receiving the information O that the event B has happened, because you also condition on receiving the information at all in the latter case. Conflating this contributes a lot to the "paradoxa" such as boy/girl paradox, Monty Hall...
![For the applicability of Bayes' theorem, it is important when the information $B$ you condition on is actually available.
Let $O \subset B$ be the event of observing $B$ with $P(O)>0$. If $O$ and $A$ are independent with respect to $P_{|B}(\cdot) = P(\cdot \mid B)$,
then
\[
P(A\mid O) = P(A \mid B).
\]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:weizynbs42yzl4y6de3z4dvu/bafkreihnkdxbvmdk4um6how7qgvurwhfhrkqhlu4baqlnkeepd7z4kkjfe@jpeg)
November 6, 2025 at 11:09 AM
The event B having happened is not the same as the receiving the information O that the event B has happened, because you also condition on receiving the information at all in the latter case. Conflating this contributes a lot to the "paradoxa" such as boy/girl paradox, Monty Hall...
Very nice by Jun Otsuka @junotk.bsky.social and Hayato Saigo: link.springer.com/article/10.1...
about causal interventions/do calculus via string diagram surgery
about causal interventions/do calculus via string diagram surgery
September 17, 2025 at 2:04 PM
Very nice by Jun Otsuka @junotk.bsky.social and Hayato Saigo: link.springer.com/article/10.1...
about causal interventions/do calculus via string diagram surgery
about causal interventions/do calculus via string diagram surgery
A golden Marburg Weidenhausen night accentuated by blue paper recycling bins…
August 8, 2025 at 7:19 PM
A golden Marburg Weidenhausen night accentuated by blue paper recycling bins…
So this is how I learned programming when I was twelve! 😃
July 14, 2025 at 5:15 PM
So this is how I learned programming when I was twelve! 😃
Yeah Moritz, *electricity*, that’s totally why you were doing this (bridges for random walks on random graphs)
June 26, 2025 at 6:14 AM
Yeah Moritz, *electricity*, that’s totally why you were doing this (bridges for random walks on random graphs)
Tomorrow at #BayesComp @rseyer.bsky.social with arxiv.org/abs/2504.12190 (poster presentation, 19 Jun 2025, 5.30pm - 7.30pm local time)
June 18, 2025 at 7:30 AM
Tomorrow at #BayesComp @rseyer.bsky.social with arxiv.org/abs/2504.12190 (poster presentation, 19 Jun 2025, 5.30pm - 7.30pm local time)
Right, you don't need error bars on error bars.
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p]. #statistics
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p]. #statistics
June 17, 2025 at 10:01 AM
Right, you don't need error bars on error bars.
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p]. #statistics
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p]. #statistics
A paper that started with a twitter conversation and brought us into very unfamiliar terrain is finally submitted with the help of new collaborator Andi Q. Wang: Compositionality in algorithms for smoothing arxiv.org/abs/2303.13865
April 19, 2025 at 6:45 AM
A paper that started with a twitter conversation and brought us into very unfamiliar terrain is finally submitted with the help of new collaborator Andi Q. Wang: Compositionality in algorithms for smoothing arxiv.org/abs/2303.13865
In German a sentence can start with any finite number of „das“:
Das “das”, das das “das”, das das “das”, das das “das” in der Basis hat, in der Basis hat, in der Basis hat, ist kaum noch zu erkennen.
Das “das”, das das “das”, das das “das”, das das “das” in der Basis hat, in der Basis hat, in der Basis hat, ist kaum noch zu erkennen.
March 8, 2025 at 10:26 AM
In German a sentence can start with any finite number of „das“:
Das “das”, das das “das”, das das “das”, das das “das” in der Basis hat, in der Basis hat, in der Basis hat, ist kaum noch zu erkennen.
Das “das”, das das “das”, das das “das”, das das “das” in der Basis hat, in der Basis hat, in der Basis hat, ist kaum noch zu erkennen.
Oh, the joy of the truth table of implication! The empty set is like a committee with no members. Each of those non-existing members is found in all other other committees. Each of them is also not found in any other committee.
January 19, 2025 at 8:30 AM
Oh, the joy of the truth table of implication! The empty set is like a committee with no members. Each of those non-existing members is found in all other other committees. Each of them is also not found in any other committee.
So it turns out we don't need to adjust, under the intervention there are no backdoor paths open #causalinference #julialang
![Screenshot:
julia> dag = digraph([U=>X, U=>Z, V=>Y, V=>Z, X=>Y, Z=>X])
DiGraph([1 => 3, 1 => 5, 2 => 4, 2 => 5, 3 => 4, 5 => 3])
julia> observed = [X, Y, Z];
julia> do!(dag, Z)
julia> dag
DiGraph([1 => 3, 2 => 4, 3 => 4, 5 => 3])
julia> adjustments = collect(list_covariate_adjustment(dag, X, Y, ∅, observed))
2-element Vector{Any}:
Set{Int64}()
Set([5])](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:weizynbs42yzl4y6de3z4dvu/bafkreigcizgc6fwacmjofij4yg4xva2ln7sb7aq2ep5zw4lp7ki5qxqr74@jpeg)
December 4, 2024 at 7:17 PM
So it turns out we don't need to adjust, under the intervention there are no backdoor paths open #causalinference #julialang
Cue the bell curve meme with “The probability is 1/6 to get a six” versus “Actually, as the die is already on the table, it is now an unknown but fix number and one is not allowed to talk about the probability to get a six anymore”
December 4, 2024 at 8:59 AM
Cue the bell curve meme with “The probability is 1/6 to get a six” versus “Actually, as the die is already on the table, it is now an unknown but fix number and one is not allowed to talk about the probability to get a six anymore”
Mom sends quality content on the family chat 😊. Here is her frozen dress modelling itself
November 24, 2024 at 9:17 AM
Mom sends quality content on the family chat 😊. Here is her frozen dress modelling itself
In any case you want to think of stochastic or deterministic transport processes `X(p)` where `p` varies.
November 22, 2024 at 9:15 AM
In any case you want to think of stochastic or deterministic transport processes `X(p)` where `p` varies.
And finally: everything is native Julia, so code is performant AND readable
November 16, 2024 at 11:30 AM
And finally: everything is native Julia, so code is performant AND readable
… and state of the art DAGITTY-like adjustment set search for causal effect estimation
November 16, 2024 at 11:30 AM
… and state of the art DAGITTY-like adjustment set search for causal effect estimation
We also have all the graphical primitives for native Julia graphs: d-separation, Bayes ball, backdoors and front doors, meta-algorithms
November 16, 2024 at 11:30 AM
We also have all the graphical primitives for native Julia graphs: d-separation, Bayes ball, backdoors and front doors, meta-algorithms
The famous Swedish Forrest Penguin (look carefully)
September 30, 2024 at 2:03 PM
The famous Swedish Forrest Penguin (look carefully)
For your next conference: how about a poster session in a ferris wheel
Photo: Dietmar Rabich / Wikimedia Commons / “Singapore (SG), View from Marina Bay Sands / CC BY-SA 4.0
Photo: Dietmar Rabich / Wikimedia Commons / “Singapore (SG), View from Marina Bay Sands / CC BY-SA 4.0
September 21, 2024 at 2:19 PM
For your next conference: how about a poster session in a ferris wheel
Photo: Dietmar Rabich / Wikimedia Commons / “Singapore (SG), View from Marina Bay Sands / CC BY-SA 4.0
Photo: Dietmar Rabich / Wikimedia Commons / “Singapore (SG), View from Marina Bay Sands / CC BY-SA 4.0
Not a lense effect but really a „smile in the sky“: Amazing #CircumzenithalArc over Gothenburg en.m.wikipedia.org/wiki/Circumz...
September 17, 2024 at 3:42 PM
Not a lense effect but really a „smile in the sky“: Amazing #CircumzenithalArc over Gothenburg en.m.wikipedia.org/wiki/Circumz...
Take an undirected cycle graph, and orient the edges by coin flip. What is the distribution of sinks? Hint: zero is a special case of an even.
November 2, 2023 at 7:54 AM
Take an undirected cycle graph, and orient the edges by coin flip. What is the distribution of sinks? Hint: zero is a special case of an even.
Let's resurrect the term "inversion"? Classically, having a prior on X and a forward model giving the conditional probability of observing Y given X, we ask for X given Y. I like the perspective of changing the direction of information flow (from X → Y to Y → X) It generalises so nicely...
October 16, 2023 at 9:13 AM
Let's resurrect the term "inversion"? Classically, having a prior on X and a forward model giving the conditional probability of observing Y given X, we ask for X given Y. I like the perspective of changing the direction of information flow (from X → Y to Y → X) It generalises so nicely...