An ensemble approach to short-term forecast of COVID-19 intensive care occupancy in Italian Regions.

摘要: The availability of intensive care beds during the Covid-19 epidemic is crucial to guarantee best possible treatment severely affected patients. In this work we show a simple strategy for short-term prediction ICU beds, that has proved very effective Italian outbreak in February May 2020. Our approach based on an optimal ensemble two methods: generalized linear mixed regression model which pools information over different areas, and area-specific non-stationary integer autoregressive methodology. Optimal weights are estimated using leave-last-out rationale. been set up validated Italy. A report its performance predicting occupancy at Regional level included.