Chris. Bart.

A map of real gdp growth in the EU

EU GDP is up 1.4% since last year, but growth is spread unevenly

Fastest: Ireland (17.1%, artifact of US pharma activity), Cyprus (3.6%), Bulgaria (3.5%), Croatia (3.4%), & Poland (3.2%)

Slowest: Luxembourg (-.4%), Finland (-.3%), Germany (-.2%), Austria (-.1%), & Hungary (.1%)

Ascaris lumbricoides in the lacrimal passage : Indian Journal of Ophthalmology An abstract is unavailable.

I found one journals.lww.com/ijo/fulltext...

The paper proposes some of an aggregated DAG. They show that it works for backdoor adjustment. The question is whether the proposed aggregated DAG can also be used to assess independences needed for front-door.

Thanks. Interesting. On a first look, it doesn't discuss SWIGs nor front-door. Both seem relevant in the context?

💡A new paper by Elias Bareinboim and Drago Plecko underscores the intractability of ignorability assumptions commonly invoked in the potential outcomes framework, explains why structural causal models—explicitly grounded in well-defined causal mechanisms—are far easier to interpret. 1/2

Stretching DuckDB w/ Common Crawl, ~1.7B rows, ~300 parquet files. ~2-3s for single-column aggregations, ~2-3 mins to SUMMARIZE the data, peaking at ~12-14GB memory usage. Not exactly real-time, but the fact you can do this on a laptop with no server setups or Spark pipelines is still amazing.

How scared should you be of “the China squeeze”? Xi Jinping masters the dark arts of the trade war

Xi Jinping masters the dark arts of the trade war

Donald Trump thinks he’s winning on trade, but America will lose The harm from tariffs will be lasting and deep

Donald Trump’s 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

This 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 host— 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

America’s tariff avalanche catches Switzerland unawares The soaring rate is based on exports of Trump’s favourite metal

The soaring rate is based on exports of Trump’s favourite metal

Ten indicators of what’s going on with America’s economy Trumponomics has begun to hit America’s growth

Donald Trump’s 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

Switzerland's tariff shock: The 39% U.S. hit no one saw coming The duty would be a "devastating" blow to the export-reliant Swiss economy and businesses, analysts say.

What did Switzerland do to be given 39% tariffs? www.cnbc.com/2025/08/01/s...

Switzerland in ‘shock’ at 39% US tariff blow New duty level on Swiss goods one of highest in the world and exceeds 31% rate outlined by Donald Trump in April

Switzerland in ‘shock’ at 39% US tariff blow https://on.ft.com/4ofunll

Sorry 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.

Now we are waiting for "mixed effects models in ... for causal inference"?

Would be great 🙂

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?

📊 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.

Heatwaves can overheat cables and stress power generation☀️

In fact, the recent heatwave in Europe may have impacted up to 15% of France’s nuclear capacity, as 17 out of 18 French nuclear facilities experienced capacity reductions 🇫🇷

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 regeneration…Simple summary

{tinytable} is a dead simple, ultra-flexible, and dependency-free #Rstats 📦 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/

The Economist is hiring a Senior Editorial Analyst Our Audience team is recruiting a specialist to assess the engagement around our journalism and drive data-modelling and statistical projects

We’re 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

has 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)

A 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

Cool. Good news.

Thanks for the information. Wondering whether there exist similar definitions outside of the US?