ATTENZIONE: Mercoledì 20 Ottobre, dalle 13:30 alle 14:00 sarà effettuato un intervento di manutenzione su agenda.infn.it. Tutte le modifiche effettuate ai contenuti, durante tale intervallo, potrebbero essere annullate alla fine dell'intervento. Si prega pertanto di anticipare o posticipare tali operazioni al di fuori dall'intervallo di manutenzione programmato.
ATTENTION: Wednesday October 20th, from 1:30pm to 2:00pm a maintenance intervention will be carried out on agenda.infn.it. All changes made to the contents, during this interval, could be canceled at the end of the intervention. Therefore, please anticipate or postpone these operations outside the scheduled maintenance interval.
16-20 September 2019
Bologna, CNAF
Europe/Rome timezone
School on Open Science Cloud

Resampling Methods

17 Sep 2019, 11:30
1h
Aula 1 (floor -1) (Dipartimento di Fisica)

Aula 1 (floor -1)

Dipartimento di Fisica

Viale Berti Pichat 6/2 40127 Bologna

Speaker

Prof. Giuliano Galimberti (University of Bologna)

Description

Choosing a proper level of complexity for a prediction rule (model selection) or evaluating its performance (model assessment) are two fundamental steps in any supervised statistical learning application. Both steps require reliable estimates of the expected prediction error. After providing some general definitions related to the prediction error in regression and classification, attention will be focused on estimating this error by re-using the observed data. In particular, methods based on validation sets, cross-validation and bootstrap will be illustrated. Advantages and disadvantages of each technique will be discussed.

Presentation Materials