Donald Trumpโs trade policy will be hard to dislodge. Its beneficiaries will be well-organised; its victims may not even realise why they are poorer than they could have been
08.08.2025 18:10 โ ๐ 343 ๐ 118 ๐ฌ 14 ๐ 4@chriba.bsky.social
Interested in data Currently pharmacometrics and causal inference Legacy: drug design, risk management, protein structures, protein folding, ... https://scholar.google.com/citations?user=R3QYvdUAAAAJ
Donald Trumpโs trade policy will be hard to dislodge. Its beneficiaries will be well-organised; its victims may not even realise why they are poorer than they could have been
08.08.2025 18:10 โ ๐ 343 ๐ 118 ๐ฌ 14 ๐ 4This baseline context has to be defined by an ominous expert. Similarly, all results of the analysis will be taken into account by the expert to identify further actions.
OR
Causal inference is challenging.
@lshtm.bsky.social and colleagues:
Any reports of Ascaris exhibiting self-replicating behavior within hostโ adult worms residing in sites like throat, nasal cavity, or tear ducts, and depositing eggs in these warm, humid microenvironments?
#Ascariasis #Parasitology #Helminths #TropicalMedicine
To the dismay of dozens of countries, the American president has returned to the madness of Liberation Day. These charts show the damage they will have to endure econ.st/4lez3Fv
04.08.2025 22:00 โ ๐ 29 ๐ 10 ๐ฌ 11 ๐ 0The soaring rate is based on exports of Trumpโs favourite metal
04.08.2025 18:30 โ ๐ 11 ๐ 4 ๐ฌ 1 ๐ 2Donald Trumpโs agenda has been consumed by his chaotic tariffs, a war on the civil service and attacks on the Fed. The dangers of his approach are starting to percolate into the economic data, as these charts illustrate
02.08.2025 18:20 โ ๐ 48 ๐ 25 ๐ฌ 9 ๐ 3What did Switzerland do to be given 39% tariffs? www.cnbc.com/2025/08/01/s...
01.08.2025 15:19 โ ๐ 0 ๐ 1 ๐ฌ 0 ๐ 0Switzerland in โshockโ at 39% US tariff blow https://on.ft.com/4ofunll
01.08.2025 09:39 โ ๐ 65 ๐ 27 ๐ฌ 16 ๐ 9Sorry guys, I know you don't like it, but every time you talk to a statistician, you realize how useless traditional stats is at conveying any idea about causal inference.
31.07.2025 08:23 โ ๐ 35 ๐ 1 ๐ฌ 3 ๐ 2Now we are waiting for "mixed effects models in ... for causal inference"?
Would be great ๐
Pinheiro and Bates 'Mixed-Effects Models in S and S-Plus', Springer, 2000. Yellow cover.
Ok so this Pinheiro and Bates book - physical copy on my shelf having survived multiple international moves and downsizings - has been my bible on mixed effects models. But it's 25 years old, is there a modern (#rstats) equivalent i should upgrade to?
26.07.2025 22:16 โ ๐ 23 ๐ 8 ๐ฌ 5 ๐ 1๐ Data update: We've just updated to the 2025 release of the OECD Government at a Glance dataset.
It has general government spending data up to 2024 and spending by function data up to 2023.
This update was led on our team by @parriagadap.bsky.social.
Point us to the text book explaining statistical methods to make causal conclusions based on all available knowledge on a topic.
11.07.2025 20:05 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Heatwaves can overheat cables and stress power generationโ๏ธ
In fact, the recent heatwave in Europe may have impacted up to 15% of Franceโs nuclear capacity, as 17 out of 18 French nuclear facilities experienced capacity reductions ๐ซ๐ท
https://loom.ly/0KDPRPY
Organ clocks for post-menopausal women taking estrogen or not taking estrogen
New organ clock data support the anti-aging benefits of estrogen for postmenopausal women. Reviewed here in context with previous and other recent reports.
erictopol.substack.com/p/new-anti-a...
T cell therapies are reshaping cancer treatment. Success may depend on boosting self-renewing, stem-like T cells. This insight also offers new strategies for treating autoimmune diseases by targeting harmful T cell regenerationโฆSimple summary
03.07.2025 14:32 โ ๐ 2 ๐ 1 ๐ฌ 0 ๐ 0{tinytable} is a dead simple, ultra-flexible, and dependency-free #Rstats ๐ฆ to turn data frames into beautiful tables: html, word, pdf, latex, typst, markdown, etc.
v0.10.0 has cool new features and important bug fixes. Check out the detailed tutorials at:
vincentarelbundock.github.io/tinytable/
Weโre looking for someone who can analyse the performance of our journalism and translate complex data into clear, jargon-free insights that support editorial decision-making. Apply by July 13th
29.06.2025 17:21 โ ๐ 18 ๐ 7 ๐ฌ 3 ๐ 1has anyone written a nice review or summary of Bayesian approaches to causal inference? most of what I'm aware of is focused on frequentist approaches, or implicitly frequentist (caring about bias, ignoring variance); and methods which may not translate easily to the Bayes world (e.g. weighting)
28.06.2025 07:12 โ ๐ 43 ๐ 8 ๐ฌ 13 ๐ 1A causal diagram. Z points to A which points to Y. There is a U that points to both A and Y. So A has unmeasured confounding, but Z is a valid instrument for A on Y
It's a musical #MEstimatorMonday (25/52) since we will focus on *instrumental* variables (IV)
IV analysis is nice because we can get around unmeasured confounding (at the cost of a more easily interpretable parameter)
we're living through one of the greatest technological revolutions in the history of mankind and it's not AI
23.06.2025 04:17 โ ๐ 6950 ๐ 2027 ๐ฌ 51 ๐ 24Cool. Good news.
14.06.2025 18:34 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Thanks for the information. Wondering whether there exist similar definitions outside of the US?
12.06.2025 20:54 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0I was so happy to see my friend, Daniel Sharfstein, quoted in this way. His work is brilliant.
12.06.2025 20:47 โ ๐ 4 ๐ 2 ๐ฌ 2 ๐ 0If you get an answer, will be interesting to see whether mixed effects models are considered as part of the solution. They should. I doubt that this has been looked at.
05.06.2025 16:34 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Is there a good paper that thinks carefully through how to apply the rubin causal modeling framework to within-subjects experiments? I always feel like I am doing a lot of hand waving when asserting causality for those designs.
05.06.2025 15:17 โ ๐ 7 ๐ 5 ๐ฌ 2 ๐ 0McSweeney's: New York Timesโ Style Guide Substitutions for "The President Violated the Constitution" www.mcsweeneys.net/articles/new...
04.06.2025 12:22 โ ๐ 2645 ๐ 821 ๐ฌ 71 ๐ 55Wishlist: explain difference between admissible adjustment set for d-separation vs. backdoor adjustment.
03.06.2025 17:30 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Over 300 views in the first day! I guess there's demand for this sort of thing. The next video (on d-Separation) will appear in about a week.
03.06.2025 16:20 โ ๐ 6 ๐ 1 ๐ฌ 1 ๐ 0Excited to learn that French-language rights for my forthcoming book have already been sold. There's a lot of French imperial history here. Details to come. www.publishersweekly.com/9781324092452
30.05.2025 13:23 โ ๐ 34 ๐ 8 ๐ฌ 2 ๐ 0