#forecasting — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #forecasting, aggregated by home.social.
-
DeepMind's WeatherNext buys a day of hurricane warning, and breaks a resolution assumption
Follow @1ban_news for daily coverage.
https://1ban.news/deepmind-weathernext-cyclone-forecasting-2026/
-
DeepMind's WeatherNext buys a day of hurricane warning, and breaks a resolution assumption
Follow @1ban_news for daily coverage.
https://1ban.news/deepmind-weathernext-cyclone-forecasting-2026/
-
DeepMind's WeatherNext model achieves breakthrough forecasting cyclones
https://deepmind.google/blog/weathernext-ai-model-achieves-breakthrough-in-forecasting-cyclones/
Comments: https://news.ycombinator.com/item?id=49220126
#HackerNews #DeepMind #WeatherNext #cyclone #forecasting #AI #breakthrough #technology
-
DeepMind's WeatherNext model achieves breakthrough forecasting cyclones
https://deepmind.google/blog/weathernext-ai-model-achieves-breakthrough-in-forecasting-cyclones/
Comments: https://news.ycombinator.com/item?id=49220126
#HackerNews #DeepMind #WeatherNext #cyclone #forecasting #AI #breakthrough #technology
-
NASA’s PUNCH Sharpens Solar Storm Forecasting in First Test
https://atlas.whatip.xyz/post.php?slug=nasas-punch-sharpens-solar-storm-forecasting-in-first-test
<p>Using continuous imagery from NASA’s PUNCH (Polarimeter to Unify the Corona and Heliosphere) mission
#forecasting #sharpens #punch #solar -
Chance? I need certainty!
-
Chance? I need certainty!
-
Just like in #forecasting and #comedy, #podcast episode releases are all about timing.
I love the #NPR #Politics Podcast, but do I even listen to this?
-
Just like in #forecasting and #comedy, #podcast episode releases are all about timing.
I love the #NPR #Politics Podcast, but do I even listen to this?
-
Watching the launch for Open Climate Fix new Netherlands solar energy forecast! ML-powered renewable energy forecasting that's been working in the UK already, now following me to NL :) Congrats folks! https://www.openclimatefix.org/ #renewables #forecasting
-
Watching the launch for Open Climate Fix new Netherlands solar energy forecast! ML-powered renewable energy forecasting that's been working in the UK already, now following me to NL :) Congrats folks! https://www.openclimatefix.org/ #renewables #forecasting
-
💝 Ah, it's so good when co-authors go through the final code review! 🛍️ Thanks Fei for correcting the forecasts! 💖
-
💝 Ah, it's so good when co-authors go through the final code review! 🛍️ Thanks Fei for correcting the forecasts! 💖
-
🌆 function forecast is used to obtain draws from the predictive density in our R packages 🌇 bsvars bsvarSIGNs bpvars bvars 🌆
🌟 Bayesian forecasting with state-of-the-art models
✨ point/density forecasting
⭐ great plots
💫 blazingly fast computations -
🌆 function forecast is used to obtain draws from the predictive density in our R packages 🌇 bsvars bsvarSIGNs bpvars bvars 🌆
🌟 Bayesian forecasting with state-of-the-art models
✨ point/density forecasting
⭐ great plots
💫 blazingly fast computations -
Getting the future right is now big business but forecasters at a Berkeley festival worry markets are corrupting the project. #predictionmarkets #philosophy #forecasting
Posted into FLIPBOARD EXCHANGE FEED 🗞️ @flipboard-exchange-feed-Econopass
-
🖤💙 oh wow! 💙🖤 have a look at all the things you can do 🖤💙 with our bpvars package for R 💙🖤 for forecasting with panel vector autoregressions 🖤💙
-
🖤💙 oh wow! 💙🖤 have a look at all the things you can do 🖤💙 with our bpvars package for R 💙🖤 for forecasting with panel vector autoregressions 🖤💙
-
💙 bpvars 🖤 is for forecasting time series data for many countries or agents 💙 this R package does that comprehensively to meet your needs 🖤 and it does the thing well! 💙
-
💙 bpvars 🖤 is for forecasting time series data for many countries or agents 💙 this R package does that comprehensively to meet your needs 🖤 and it does the thing well! 💙
-
💘 interested in super precise and reliable forecasts? The new R package bvars is the way to go!
👾 great tools for forecasting!
👽 super fast routines!
👻 models proven best at forecasting
🌐 https://cloud.r-project.org/package=bvars
(where's the astronaut emoji when you need it?) -
💘 interested in super precise and reliable forecasts? The new R package bvars is the way to go!
👾 great tools for forecasting!
👽 super fast routines!
👻 models proven best at forecasting
🌐 https://cloud.r-project.org/package=bvars
(where's the astronaut emoji when you need it?) -
💘 the new R package bvars includes state-of-the-art forecasting models 🚀
🔭 They are most useful for macroeconomic and financial forecasting!
🤖 the range of models is already great! ...and more to come!
-
💘 the new R package bvars includes state-of-the-art forecasting models 🚀
🔭 They are most useful for macroeconomic and financial forecasting!
🤖 the range of models is already great! ...and more to come!
-
Study Highlights Growing Importance Of Multi-Day Storms In Future U.S. Flood Risk
--
https://news.okstate.edu/articles/engineering-architecture-technology/2026/study-highlights-growing-importance-of-multi-day-storms-in-future-u.s.-flood-risk <-- shared technical article
--
https://doi.org/10.1088/2752-5295/ae4f14 <-- shared paper
--
“Extreme rainfall is projected to intensify as the climate warms, yet whether the greatest increases will occur in multi-day or single-day events remains uncertain. This knowledge gap is particularly pressing given recent catastrophic floods triggered by multi-day rainfall events, prompting the question of whether multi-day events could, in fact, intensify more than their daily counterparts, and by how much. This study addresses this question using an ensemble of 34 downscaled Earth System Models under two Shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5), focusing on changes in extreme rainfall by the end of the century across ten regions of the contiguous United States. [Their] statistical framework evaluates model agreement, ensemble-mean changes, and the significance of these changes for both daily and multi-day rainfall extremes. Results show that extreme rainfall amounts are expected to increase for most regions and durations. The degree of intensification, however, depends strongly on event rarity and regional climate characteristics. Notably, in the U.S. western Gulf Coast region, very rare multi-day events (e.g., 500 year return period) are projected to intensify more than their daily counterparts, a phenomenon that could be explained by increased stalling of tropical cyclones, which can prolong heavy rainfall over multiple days. These results challenge the assumption that daily extremes dominate future risk and highlight the need to consider event duration when updating flood-hazard maps, design standards, and adaptation planning…”
#Flooding #FloodRisk #FloodInsurance #FloodAwareness #Explore #FloodPreparedness #FlashFlooding #ClimateResilience #climatechange #extremeweather #DisasterPreparedness #StormwaterManagement #FloodSafety #CommunityResilience #risk #hazard #model #modeling #floodrisk #multiday #rainfall #precipitation #storm #water #hydrology #hydrography #planning #policy #regulations #climatemodel #CONUS #USA #publicsafety #cost #economics #damage #loss #infrastructure #spatiotemporal #spatialanalysis #earthsystemmodels #forecasting #meteorology #designstandards #floodmapping #mitigation #flood -
Study Highlights Growing Importance Of Multi-Day Storms In Future U.S. Flood Risk
--
https://news.okstate.edu/articles/engineering-architecture-technology/2026/study-highlights-growing-importance-of-multi-day-storms-in-future-u.s.-flood-risk <-- shared technical article
--
https://doi.org/10.1088/2752-5295/ae4f14 <-- shared paper
--
“Extreme rainfall is projected to intensify as the climate warms, yet whether the greatest increases will occur in multi-day or single-day events remains uncertain. This knowledge gap is particularly pressing given recent catastrophic floods triggered by multi-day rainfall events, prompting the question of whether multi-day events could, in fact, intensify more than their daily counterparts, and by how much. This study addresses this question using an ensemble of 34 downscaled Earth System Models under two Shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5), focusing on changes in extreme rainfall by the end of the century across ten regions of the contiguous United States. [Their] statistical framework evaluates model agreement, ensemble-mean changes, and the significance of these changes for both daily and multi-day rainfall extremes. Results show that extreme rainfall amounts are expected to increase for most regions and durations. The degree of intensification, however, depends strongly on event rarity and regional climate characteristics. Notably, in the U.S. western Gulf Coast region, very rare multi-day events (e.g., 500 year return period) are projected to intensify more than their daily counterparts, a phenomenon that could be explained by increased stalling of tropical cyclones, which can prolong heavy rainfall over multiple days. These results challenge the assumption that daily extremes dominate future risk and highlight the need to consider event duration when updating flood-hazard maps, design standards, and adaptation planning…”
#Flooding #FloodRisk #FloodInsurance #FloodAwareness #Explore #FloodPreparedness #FlashFlooding #ClimateResilience #climatechange #extremeweather #DisasterPreparedness #StormwaterManagement #FloodSafety #CommunityResilience #risk #hazard #model #modeling #floodrisk #multiday #rainfall #precipitation #storm #water #hydrology #hydrography #planning #policy #regulations #climatemodel #CONUS #USA #publicsafety #cost #economics #damage #loss #infrastructure #spatiotemporal #spatialanalysis #earthsystemmodels #forecasting #meteorology #designstandards #floodmapping #mitigation #flood -
Bonkus, just bonkus… I have looked at three weather apps (including BOM) and all of them say it’s raining here, yet a cursory look out my windows reveals blue skies for miles around. Not one tiny shadow of a cloud. Save for thos on the horizon (what 40 kms away? More? I’m on a hill).
Absolutely crap weather forecasting. Some apps show my area surrounded by lightning and dense cloud cover. Heaven help me the sattelite ovelay is also showing the whole countryside under thick cloud cover. WTF?
Anytime now, I’ll come awake and see the garden drenched under thundering rain…
-
Bonkus, just bonkus… I have looked at three weather apps (including BOM) and all of them say it’s raining here, yet a cursory look out my windows reveals blue skies for miles around. Not one tiny shadow of a cloud. Save for thos on the horizon (what 40 kms away? More? I’m on a hill).
Absolutely crap weather forecasting. Some apps show my area surrounded by lightning and dense cloud cover. Heaven help me the sattelite ovelay is also showing the whole countryside under thick cloud cover. WTF?
Anytime now, I’ll come awake and see the garden drenched under thundering rain…
-
📈 ¿Modelo simple o complejo? La clave no es la fórmula, es el contexto.
Elegir el mejor modelo de pronóstico puede parecer un rompecabezas. En este video te comparto un enfoque práctico para tomar decisiones más acertadas.
🎯 Lo que verás:
• La métrica correcta para comparar modelos (no todo es el R²).
• Cómo evitar el sobreajuste (y por qué un modelo simple puede ganar).#Forecasting #ModelosDePronóstico #DataScience #MachineLearning #EstadísticaAplicada #MastodonCiencia
-
📈 ¿Modelo simple o complejo? La clave no es la fórmula, es el contexto.
Elegir el mejor modelo de pronóstico puede parecer un rompecabezas. En este video te comparto un enfoque práctico para tomar decisiones más acertadas.
🎯 Lo que verás:
• La métrica correcta para comparar modelos (no todo es el R²).
• Cómo evitar el sobreajuste (y por qué un modelo simple puede ganar).#Forecasting #ModelosDePronóstico #DataScience #MachineLearning #EstadísticaAplicada #MastodonCiencia
-
Here is a curious idea: if we can somehow estimate the importance of trend/seasonal components for your data, you can use this in model building and forecasting. But how can we do this first step? Hans Levenbach has an answer with his simple EDA technique. Let me explain: https://openforecast.org/2026/05/18/hans-levenbach-s-classification-scheme-for-trend-seasonal-components/
-
Here is a curious idea: if we can somehow estimate the importance of trend/seasonal components for your data, you can use this in model building and forecasting. But how can we do this first step? Hans Levenbach has an answer with his simple EDA technique. Let me explain: https://openforecast.org/2026/05/18/hans-levenbach-s-classification-scheme-for-trend-seasonal-components/