Distributed Sequential Zero-Inflated Poisson Regression

0549265 - ÚTIA 2022 CZ eng V - Výzkumná zpráva
Žemlička, R. - Dedecius, Kamil
Distributed Sequential Zero-Inflated Poisson Regression.
Praha: ÚTIA AV ČR, v. v. i.,, 2021. 11 s. Research Report, 2393.
Institucionální podpora: RVO:67985556
Klíčová slova: Poisson regression * zero inflation * GLM
Obor OECD: Applied mathematics
http://library.utia.cas.cz/separaty/2021/AS/dedecius-0549265.pdf

The zero-inflated Poisson regression model is a generalized linear model (GLM) for non-negative count variables with an excessive number of zeros. This letter proposes its low-cost distributed sequential inference from streaming data in networks with information diffusion. The model is viewed as a probabilistic mixture of a Poisson and a zero-located Dirac component, whose probabilities are estimated using a quasi-Bayesian procedure. The regression coefficients are inferred by means of a weighted Bayesian update. The network nodes share their posterior distributions using the diffusion protocol.

Trvalý link: http://hdl.handle.net/11104/0325721