A latent variable model for identifying and characterizing food adulteration Recently, growing consumer awareness of food quality and sustainability has led to a rising demand for effective food authentication methods. Vibrational spectroscopy techniques have emerged as a prom...

Hot off the press. A preprint by @alecasa.bsky.social, Michael Fop and I. A new model for characterising and identifying adulteration in food authenticity studies. arxiv.org/abs/2512.13939

link 📈🤖
A latent variable model for identifying and characterizing food adulteration (Casa, Murphy, Fop) Recently, growing consumer awareness of food quality and sustainability has led to a rising demand for effective food authentication methods. Vibrational spectroscopy techniques have emerged a

@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...

Dati, economia e sfide reali. La prima edizione del Data4Econ Hackathon Il 3 giugno si terrà la prima edizione dell’unibz Data4Econ Hackathon, una competizione dove si incontrano analisi statistica, data science ed economia.

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.”

👏🏼