Pinned Tweet
My 2.5 hour tutorial at ACML2020 on Forecasting for Data Scientists is now online: cbergmeir.com/talks/acml-tut…
All the way since the beginnings of my PhD I tried to get my head around how to evaluate time series forecasting methods properly. Our latest go at the topic is a quite comprehensive tutorial with some guidance on best (and worst) practices, here: arxiv.org/abs/2203.10716
We just published a press release about our quite successful collaboration on short-term wind and solar power forecasting with Worley.
monash.edu/news/articles/ai-…
3
We've just launched the IEEE-CIS Technical Challenge on Predict+Optimize for Renewable Energy Scheduling, with US$20k in prize money. Forecast energy demand and solar power for buildings on a Monash campus to optimally schedule lectures and batteries.
ieee-dataport.org/competitio…
We have put together a new time series data repository for forecasting. It is dedicated to sets of series for cross-learning/global modelling, and to single but very long series, find it here:
Paper link: lnkd.in/gHJH5CB
Website: lnkd.in/g5WABb6
Christoph Bergmeir retweeted
Applications for our 2021 #Women in #Technology #Alumni#Mentoring program are officially open!
If you're a #tech professional eager to provide a three-month, one-to-one mentorship to a driven woman #IT student, apply now: ow.ly/ba1h50Dlnru.
Our two final papers for 2020 are out, a simulation study to shed some light into "relatedness" for global models arxiv.org/abs/2012.12485 and a study around how to localise global models with ensembling strategies arxiv.org/abs/2012.15059
Christoph Bergmeir retweeted
Me, teaming up with Hansika, Rakshitha from the Monash University, Australia was able to achieve the overall 4th position according to the final ranking of the IEEE scientific committee. @CBergmeir @MonashInfotech #Forecasting #Energy #IEEECIS
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
Christoph Bergmeir retweeted
Our article, "Forecasting: theory and practice", is the first encyclopedic-type approach to the field of forecasting, covering many different methods and applications.
My huge thanks to my many co-authors for making this possible.
Download here: yanfei.site/docs/paper/ftp.p…
Neural Prophet that we develop with Stanford University and Facebook just got released
Producing high quality forecasting is not an easy problem for machines or analysts. We’ve built NeuralProphet using #deeplearning to do time series modeling, and help solve this. It’s a PyTorch-based version of our Prophet forecasting package. Code: github.com/ourownstory/neura…
Check our invited session on Global Forecasting Models at Virtual ISF2020 next Monday whova.com/embedded/session/i…
1
We have a new paper on forecasting weekly time series: arxiv.org/abs/2010.08158






