@alecasa.bsky.social and Davide Ferrari (@unibz.bsky.social) tackle the problem of selecting the number of components in Gaussian mixture models, where traditional single-model criteria like BIC often overlook model selection uncertainty. Learn more⤵️ www.sciencedirect.com/science/arti...

I partecipanti, suddivisi in team, lavoreranno su dataset reali forniti da #Alperia e #GruppoDolomitiEnergia (sponsor dell’evento), sviluppando modelli predittivi e interpretazioni economiche innovative. @alecasa.bsky.social Leggi di più: www.unibzmagazine.it/en/magazine/...

link 📈🤖 Confidence set for mixture order selection (Casa, Ferrari) A fundamental challenge in the application of finite mixture models is selecting the number of mixture components, also known as order. Traditional approaches rely on selecting a single best model using information criteria. Howev

First time using Bluesky to advertise a new preprint posted on arXiv! So here it is: a clever and more robust workaround to the single-best-model paradigm when selecting the number of components in the mixture modeling framework.

“L’attuale orientamento di X non è compatibile con i valori fondamentali di un’università come la Libera Università di Bolzano, basati su fatti, trasparenza e apertura verso il mondo.” 👏🏼