Discover and read the best of Twitter Threads about #forecasting

Most recents (24)

Are Transformers useless for time series forecasting? NO! We showcase how Transformers achieve SOTA results, beating DLinear models.

Here's a blog post introducing the Autoformer model and how effective it can be at #timeseries #forecasting: Image
The paper "Are Transformers Effective for Time Series Forecasting?", claims that a simple linear model DLinear outperforms Transformer based models by a large margin.
Why does that happen in their case? Firstly, they compare a linear model with massive models, and they even acknowledge it. Secondly, they compare their univariate model with multivariate versions of the Transformers, which on the small datasets gives poor performance. Image
Read 6 tweets
MA businesses + companies #discover & #SAVEMONEY with Brian Plain's 25 ways Marlborough & Boston MA area business can save money + #reducecosts & corporate expenditures:

1. Embrace cost reduction strategies to optimize your business operations. #ReduceCosts #BusinessOptimization
2. Focus on delivering exceptional service to your customers while minimizing expenses. #ServiceExcellence #CostReduction

3. Cultivate an agile workforce that can adapt to changing market demands and drive cost savings. #AgileWorkforce #ReduceCosts
4. Invest in team development to enhance skills and productivity, leading to efficient cost management. #TeamDevelopment #CostManagement

5. Implement streamlined systems and processes to improve operational efficiency and reduce costs. #Systems #ProcessImprovement #CostReduction
Read 15 tweets
Learning data science on your own is tough...

...(ahem, it took me 6 years)

So here's some help.

5 Free Books to Cut Your Time In HALF.

Let's go! 🧵

#datascience #rstats #R Image
1. Mastering #Spark with #R

This book solves an important problem- what happens when your data gets too big?

For example, analyzing 100,000,000 time series.

You can do it in R with the tools covered in this book.

Website: Image
2. Geocomputation with #R

Interested in #Geospatial Analysis?

This book is my go-to resource for all things geospatial.

This book covers:
-Making Maps
-Working with Spatial Data
-Applications (Transportation, Geomarketing)

Website: Image
Read 8 tweets
🎉 We are thrilled to announce the release of the latest version of mlforecast a #Python library for Scalable #machinelearning 🤖 for #timeseries #forecasting

🚀 This version comes with exciting new features that are sure to make forecasting even more efficient and accurate

🔮 Conformal Prediction: We've added the ability to generate probabilistic forecasts using conformal prediction. You can choose the levels of your intervals and the number of conformity scores to use.
⚡️ In addition, mlforecast includes the capability to fit multiple time series with global models. That means you don’t need to calculate sequential conformal intervals for each series and period. I.e., conformal prediction for time series just got a lot faster and easier. 🤯
Read 8 tweets
🎉 We are very excited to release the new features of NeuralForecast! 🥳🚀

With this release time series forecasting with neural models is even more accessible and powerful. Here are some of the highlights


#timeseries #python #deeplearning #forecasting
📈 New Models:
- Temporal Convolution Network
- AutoTFT (Transformers)
🧮 Recurrent models (RNN, LSTM, GRU, DilatedRNN) can now take static, historical, and future exogenous variables.
Read 9 tweets
Multi-temporal variability forecast of particulate organic carbon in the Indonesian seas


This research aimed to forecast the temporal variability of POC in Indonesian waters.
POC forecasting is essential in predicting the productivity of Indonesian seas, for several reasons, namely, that it will affect marine productivity, the Indonesian seas are dynamic, and the existence of global threats to the ocean (climate change, ocean acidification, and d'Ox)
Read 15 tweets
Want to see the power of Time Series forecasting in action? 📈

Let's take a look at how the retail giant Walmart uses it to improve its operations. 🤯


#timeseries #forecasting #retail #example Image
Walmart has been using time series forecasting for decades to predict sales and optimise inventory levels.

By analysing historical sales data, weather patterns, and other factors, they're able to make more accurate predictions about future demand for products.
This allows them to ensure they have the right products in stock at the right time, reducing stockouts and increasing customer satisfaction.
Read 8 tweets
1/2 "Here, we demonstrate an increased survival of infected #honey #bee #larvae after their #queen was #vaccinated, compared to offspring of #control #queens (#placebo vaccinated)."…
2/2 "These results indicate that #trans-#generational #immune #priming (#TGIP) in insects can be used to majorly enhance #colony #health, protect #commercial #pollinators from #deadly #diseases, and reduce high #financial and #material losses to #beekeepers"
1/2 "The idea came to be known as the “#butterfly #effect” after Lorenz suggested that the flap of a butterfly's wings might ultimately cause a tornado."…
Read 4 tweets
Thanks ICP for publishing this essay on #Ritamic Decision Policy. Into year-3 of this series, this post summarizes manthan, research and study of the Vedic origins of sustainable decision policy and strategy, with contemporary examples all can relate to.
Post focuses on sustainable policy to make not one, but a series of interconnected decisions over time.

Can ideas of Vedanta be applied here?
short ans: Y.

Content is of interest to those making gov or private policy, designers, startups, #Ganita/stem students, young parents.
The post is divided into 7 sections with links to each at the top. Those who simply want a easy-to-remember idea of Ritamic decision policy can go to the examples- read how Air India altered its DEL-SFO route in harmony with Ritam.…
Read 38 tweets
Conformal prediction for #timeseries #forecasting is the hottest 🔥🔥🔥🔥 🚀🚀🚀🚀 area of research and applications right now.

The whole story started in summer 2021 when researchers from Georgia Tech published the first ever paper applying conformal prediction to time series.
and presenting it at the first event at major conference totally dedicated to conformal prediction (ICML DFUQ 2021 organised by @ml_angelopoulos and @stats_stephen
Roll forward just 1.5 yeast - there are multiple papers published on the subject, several #opensoure libraries like MAPIE implemented the original first ever paper from Georgia Tech (method called EnbPI).
Read 7 tweets
Continuing #LeBronWatch. The @Lakers have now played 7 games. How has #LeBronJames' performance changed our initial forecasts -…
Answers in thread
The possibility of LeBron getting to Kareem's mark by game 49 has gone up. Currently, the most likely range is game 48 to 52, with 80% of the results falling in that range. We are 94% sure that LeBron gets there by game 52.
There are very small chances that LeBron's quest extends into games 55, 56, or 57. Similarly, a small set of simulations (<2 percent) point to games 44 and 45.
Read 7 tweets
Game one has come and gone. Here is how the predictions have shifted for LeBron eclipsing Kareem.
We first made these predictions here -…
Time to update the forecast. #ContinuousForecasting
Lebron scored 31 points in the first game of the season. 31 is a better score than 70% of the games Lebron played in the last three seasons.
Our New "LeBron plays all games" forecast -
Read 8 tweets
I'm excited about this next session at #ExploreSAS => "Super Demo: The Science Behind the Batting Lab" by @SASsoftware <= #DataLiteracy At Bat…
#SportsAnalytics #BigData #Analytics #AI #MachineLearning #DataScience #SASVisionary #Data4All #Analytics4All Image
Live session at #ExploreSAS => "Super Demo: The Science Behind the Batting Lab" by Gunce Walton,
Manager for Advanced Analytics R&D at @SASsoftware
#DataLiteracy #SportsAnalytics #BigData #Analytics #AI #MachineLearning #DataScience #SASVisionary #Data4All #Analytics4All ImageImageImageImage
Read 5 tweets
When it comes to Time Series, colleges and universities have it all wrong.

A time series thread 🧵

#rstats #excel #python #timeseries
Universities are stuck in the past, teaching ARIMA.

But the cold reality is that ARIMA is NOT winning time series competitions & ARIMA is NOT helping companies solve BIG forecasting problems.
To be frank, ARIMA is too slow.

When you use ARIMA, you fall into a trap. You think, hey, this is what they're teaching me...

It must be good, right?
Read 12 tweets
Howard Marks en son yazdığı “memo”sunda makroekonomik tahmin yapmanın gerekliliğini sorguluyor.

Bir tahmin yapıyorsak bazı varsayımlara göre yapıyoruzdur. Yoksa atmasyon bir şey olur. Yapılan her varsayım ise bizi gerçeklikten uzaklaştırıyor.
Tahmin yapmak, normalde hayli karmaşık etkileşim içinde olan bir sürü veriye, gelişmeye aynı anda hakim olmayı gerektiriyor. Pozitif bir bilim olmayan ve insan davranışlarına oldukça duyarlı iktisatta ise doğru tahmin yapmak neredeyse imkansızlaşıyor.
Geleceği tahmin ederken geçmiş iyi bir rehber midir, yani her bir durum karşısında hep aynı şeyler mi olur (stationarity)? Pozitif bilimlerde bu geçerli olabilir. Ama iktisatta hep düşünmediğimiz bir şeyler mutlaka vardır ve her zaman beklenmedik bir şey olur.
Read 11 tweets
Another great development from @nixtlainc, ETS from StatsForecast is much more accurate and over 100 faster than NeuralProphet.

nice work @fede_gr and @nixtlainc team

#timeseries #forecasting
Last year a few data scientists including myself exposed severe issues with facebook prophet resulting in facebook pulling down claims such as “with facebook prophet anyone can produce better forecasts than human experts “.
Still the hope was that with Neural Prophet things will be better and one can at least partially repair the worst forecasting model of the XXIst century - Facebook Prophet.…
Read 8 tweets
You might hear a geologist say: "We can't predict earthquakes."

Well, WHY NOT???



#earthquake #earthquakeprediction
First, let's define "prediction". A useful #earthquakeprediction will tell you where, when, and how big a significant #earthquake will be, with a reasonably high success rate.

That's different from #forecasting, #earlywarning, and #aftershocks. 2/n

To date NO ONE has developed an effective way to predict significant earthquakes.

Issue #1: Faults are fractal.

"But we know where the tectonic plates are!"

Yes, mostly, but plate boundaries are complicated - see the maps below comparing plates vs. faults. 3/n

Read 23 tweets
Bayern sagte im Dez. 2.000 neue Nachverfolger zu, 15 pro 100k Einwohner. Als Zielzahl des Bundes gelten 25/100k. Die Fluktuation war wohl zu groß.

Frage: Weiß jemand wie die Lage heute ist?

Wiesbaden, Anfang März, sollen "statt ursprünglich 5 nun bis zu 10 Kontaktnachverfolger pro 20 000" zur Verfügung stehen. Was ist die Realität? Weiß dass jemand, liebe #Hessen?…
Kreis & Stadt Würzburg hatten am 29.12. 59 #Nachverfolger auf 280k Einwohner, 21 pro 100k.

Die Zielzahl des Bundes ist 25/100k. Die USA planten min. 30/100k, in Wuhan konnten 80/100k die Lage lösen.

Weiß jemand wie es heute ist #Würzburg?…
Read 44 tweets
Daily Bookmarks to GAVNet 12/04/2020…
Preserving a Sense of Wonder in DNA - Issue 92: Frontiers - Nautilus…

The ocean in humanity’s future…

#future #humanity #ocean
Read 8 tweets
Daily Bookmarks to GAVNet 12/02/2020…
Preserving a Sense of Wonder in DNA - Issue 92: Frontiers - Nautilus…

The ocean in humanity’s future…

#future #humanity #ocean
Read 8 tweets
🌄Good evening everyone!

Welcome to In Conversation with 𝐃𝐫 𝐑 𝐂𝐡𝐢𝐝𝐚𝐦𝐛𝐚𝐫𝐚𝐦, Former Principal Scientific Adviser to Govt. of India, Chairman AEC & Secretary @DAEIndia👥

🗣️Please post your recommendations, comments with #ChatwithRChidambaram
👸👸This live-tweet thread is curated by @AnamZille & @Jengovz on behalf of @spf_in

🌸🌸Opening the session with a warm welcome by @ChagunBasha @DSTCPRIISc @PrinSciAdvGoI
Read 43 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!