Mike

Dr. en Innovación y Gobernanza para el Desarrollo Sustentable #naturebasedsolutions

[Joachim Schork @JoachimSchork]

Before estimating a treatment effect, there is one important question:

Are the treatment and control groups actually comparable?

When working with observational data, the answer is often no. Individuals who receive a treatment frequently differ from those who do not, making direct comparisons potentially misleading.

Propensity Score Matching (PSM) is a widely used method to address this problem. By matching treated and untreated individuals with similar propensity scores, it creates groups that are more comparable with respect to their observed characteristics.

The figure below compares the distribution of propensity scores before and after matching. After matching, the treatment and control groups show much greater overlap, indicating that the matched individuals are more comparable than they were originally.

However, there is an important limitation. PSM can only balance observed covariates. If important confounders are unobserved or unavailable, they may still bias the estimated treatment effect, even when the matching appears successful.

Don't miss future tutorials on statistics, data science, AI, and programming with R and Python. Subscribe to my newsletter.

Learn more: https://t.co/ktUcWo9XpO

#Statistics #R4DS #DataAnalytics #Python #RStats

[Carolina Espinosa Luna @espinosacrim]

Acumular lecturas no es ampliar el estado de la cuestión.

[Philipp Heimberger @heimbergecon]

We have a new paper on the fiscal consequences of climate change: global temperature shocks substantially reduce economic output and tax revenue, and raise government debt. Thread 👇

[nature @Nature]

Nature research paper: Amazon carbon emissions driven by human actions amplified by droughts

https://t.co/NoX9JD9aE2

[Jose Ramos Vivas @joseramosvivas]

☀️🏃‍♂️ ¿Entrenar por la MAÑANA o por la NOCHE? 🌙🏋️‍♀️😴

🔬📚 Un reciente metaanálisis publicado en Sleep Medicine Reviews analiza 18 ensayos clínicos aleatorizados para evaluar cómo influye el horario del ejercicio físico en el sueño. 💤🧠

🔍✨ PRINCIPALES HALLAZGOS:

☀️🌙 1️⃣ Mañana frente a noche:
En general, la evidencia disponible no muestra una ventaja clara del ejercicio matutino frente al vespertino/nocturno en la mayoría de los parámetros del sueño. ⚖️😴

🌙⏱️ 2️⃣ Continuidad del sueño:
El ejercicio nocturno se asoció con un ligero aumento de la vigilia tras el inicio del sueño (WASO). 🛌📊

⚠️ El efecto fue pequeño y debe interpretarse con cautela debido a posibles efectos de estudios pequeños. 🔎

🏋️‍♂️🔥 3️⃣ La intensidad importa:
Las diferencias relacionadas con el horario parecen ser más pronunciadas cuando el ejercicio se realiza a intensidades elevadas. 💪📈

💡🎯 CONCLUSIONES

✅ El mejor horario para entrenar puede ser aquel que favorezca la constancia y la adherencia, respetando las preferencias y circunstancias individuales. 🏃‍♀️🕒

🌙⚠️ No obstante, el ejercicio intenso por la noche podría afectar ligeramente a la continuidad del sueño en determinadas circunstancias. 💤

🧪🔬 ¿CÓMO MEJORAR FUTURAS INVESTIGACIONES?

🧠⏰ Controlar mejor los cronotipos individuales (personas más matutinas o vespertinas).

🛏️🌡️ Estandarizar las condiciones ambientales del descanso.

📊🔍 Analizar con mayor precisión cómo interactúan el horario, la intensidad del ejercicio y los ritmos circadianos.

🔗📖 ESTUDIO COMPLETO 👇🏻
https://t.co/tZmhwKIfWM

#Sueño #Ejercicio #MedicinaDelSueño #Cronobiología #Entrenamiento #ActividadFísica #RitmosCircadianos #Salud #SleepScience #Ciencia

[Kirk Borne @KirkDBorne]

Find the GitHub Code Repository for “Causal Inference and Discovery in Python” at https://t.co/MJWLYrSRnU via @PacktDataML
+
Get the book here: https://t.co/Q1aU74Va2L

[Carolina Espinosa Luna @espinosacrim]

La UNAM ha establecido que la evaluación académica debe incorporar la perspectiva de género. Sin embargo, los nuevos criterios para acceder al Programa de Primas al Desempeño (PRIDE D) no precisan cómo hacerlo.
Aquí comparto mi opión al respecto.
https://t.co/DlUVoC6tvy

[Global Change Biology @GlobalChangeBio]

RESEARCH ARTICLE
Silent Alarms? Unveiling Destabilization Risk in Terrestrial Ecosystem Carbon Fluxes
🔗 https://t.co/xD32IGYpFA

[Kirk Borne @KirkDBorne]

Mental Models — How understanding the mind can transform the way you work and learn: https://t.co/C0vHisONp3

[USGS Landsat @USGSLandsat]

Precision matters, even in 47-year-old data. 🛰️

Several Landsat 2 and Landsat 3 Multispectral Sensor (MSS) products were recently reprocessed to correct brightness value errors first identified in early 2025. About 44,000 of these products were originally processed with incorrect gain and offset values, specifically in Band 4 and Band 5, which ended up looking brighter or darker than they should have.

This reprocessing also fixed about 1,700 MSS products that had a second problem. A processing bug caused about 27,000 extra pixels per scene, roughly 22% of each image, to be mistakenly marked as cloud, mostly along cloud edges.

It's a small fix with a big purpose.

Correcting these issues reflects our ongoing commitment to providing high-quality data to the Landsat user community. Keeping decades-old imagery accurate matters just as much as collecting new data. It's what allows Landsat's 50+ year archive to remain a reliable resource for studying long-term change on Earth's surface.

Reprocessed products are ready to download from USGS Landsat Data Access: https://t.co/hXVSO468s3

📸: This Landsat 3 MSS scene over the Los Angeles, California, area was reprocessed to correct brightness value errors. Smoke from wildfires can be seen rising from the Angeles National Forest in September 1979.
Credit: USGS

#Landsat #RemoteSensing

[Global Change Biology @GlobalChangeBio]

SCIENCE BEHIND THE NEWS
“Wildlife and Climate” Scientific Consensus on the Roles of Wild Animals in the Global Carbon Cycle and Natural Climate Solutions
🔗 https://t.co/jvhJ9vq8uM

[Carolina Espinosa Luna @espinosacrim]

Cuando una universidad mantiene durante años en la incertidumbre laboral a quienes están comenzando su carrera académica, compromete su propia capacidad de consolidar el conocimiento que esas personas producen.

[Virdika R. Utama @virdikaa]

Ini salah satu paper terbaik yang gue baca tahun ini. Salah satu penulisnya adalah profesor gue: Dylan Loh. Plus, Ayse Zarakol ini salah satu penulis favorit gue. Bukunya Before the West is another level sih.

https://t.co/lhJD5suXJA

[Amir Ghasemi 📱 @amir_ghm]

خواهرا و داداشام سفت بشینین؛ بعد از Adobe، حالا نوبت Microsoft Office شده 😂

همون تیم برای Word، PowerPoint و Excel هم جایگزین رایگان و اوپن‌سورس ساخته و گذاشته برای دانلود. حدود ۷۰ تا ۸۰ درصد کاراییشون رو پیاده کرده

اساتید، سوت پایانو بزنیم یا زوده؟ چی یاد بگیریم، یا چی بزنیم که دو روز بعد نسخه رایگانش نیاد؟ 🫠

ازینجا نصبشون کنین:
https://t.co/wMOgGok6Cy

[System Dynamics Society @systemdynamics_]

🚀 JOIN us for a FREE webinar: “Combining #SystemDynamics with Discrete Event and Agent-Based Methods: Why and How” with Andrei Borshchev (The AnyLogic Company)

📅 November 3 @ 11:00 am - 12:00 pm NY Time
🔗 RSVP: https://t.co/ZUFqNhDn5P

#systemsthinking #SeminarSeries

[silvan @silvanrec]

🚨 DEJA DE PAGAR PHOTOSHOP... ESTA ALTERNATIVA ES GRATIS Y PESA SOLO 11 MB

Un desarrollador se cansó de pagar Photoshop y creó su propia alternativa.

Es open source: cualquiera puede descargarlo, modificarlo y adaptarlo a su forma de trabajar.

Se llama Compositor y ya supera las 8.000 estrellas en GitHub.

Tiene lo que más usas de Photoshop:

→ Capas, máscaras y modos de fusión
→ Relleno según el contenido
→ Seleccionar sujeto y quitar fondo
→ Filtro Camera Raw
→ Pincel corrector y tampón de clonar

Sin suscripción. Sin pagarle a Adobe cada mes.

Te dejo el repo abajo ��

[Lectócratas @lectocratas]

📚 LA VENGANZA DE LA GEOGRAFÍA
Cómo los mapas condicionan el destino de las naciones

✍️ Robert D. Kaplan (2012)

📥 DESCARGA GRATUITA:
https://t.co/ZRkwTw0qwU

Lectócratas | Lectura Política

[PNASNews @PNASNews]

Researchers have identified various potential climate “tipping points,” but some, such as the collapse of the Greenland Ice Sheet, get most of the attention, while others, such as the abrupt thawing of boreal permafrost, are less studied. In PNAS: https://t.co/O54P8v4Qze

[Bruno César Capucin @BrunoCapucin]

O padrão climático observado nos trópicos em setembro reflete a forte influência do El Niño em andamento. A convecção intensa sobre as águas anomalamente quentes do Pacífico Equatorial favorece movimentos ascendentes mais vigorosos nessa região, acompanhados por subsidência compensatória sobre a Oceania e partes do Brasil tropical.

Sobre a bacia Amazônica, essa circulação anômala tende a dificultar o estabelecimento e a evolução da estação chuvosa, favorecendo precipitações abaixo da climatologia em diversas áreas. À medida que o El Niño se aproxima de seu pico, esperado para o último trimestre do ano, a intensificação dos movimentos ascendentes no Pacífico deverá reforçar essas zonas de subsidência.

A convecção anômala no Pacífico Equatorial também interfere na circulação extratropical, modulando a posição e a intensidade da corrente de jato, além de manter ondas planetárias como pano de fundo da circulação de grande escala (teleconexão atmosférica).

Em setembro de 2026, o Jato Subtropical sobre a América do Sul permaneceu mais ao norte do que o normal, inclusive em relação ao padrão climatológico típico de anos de El Niño. Essa configuração ajuda a explicar os episódios de severidade no Sul e as chuvas extremas em áreas do Sudeste do Brasil.

Atualmente, a temperatura absoluta da superfície do mar na região do Niño 3.4 é de 29,8 °C, valor igual ao observado em meados de novembro de 2015, quando aquele forte El Niño atingia seu pico. Mais do que uma comparação direta entre os dois eventos, o dado chama atenção para a rápida evolução do aquecimento no Pacífico Equatorial em 2026, especialmente considerando que o fenômeno ainda se aproxima do período climatológico de máxima intensidade.

[Sven Nyholm @SvenNyholm]

I reply to Moti Mizrahi's generous review of my book *The Ethics of Artificial Intelligence: A Philosophical Introduction*, here: https://t.co/JiuM90MR7q Thanks to the Social Epistemology Review and Reply Collective for hosting this back-and-forth! #AIEthics #philosophy