In this paper we focus on the use of large data sets to implement optimal designs. Our final goal is an experimental design that is optimal by means of sequential designs. In this work the improvement is related to two aspects: weights and stopping rule. The definition of weights is relevant to discriminate between more than two models. The settlement of a stopping rule is important to evaluate the contribution of the new selected experimental point. The main problems here faced are the choice of the best model according to the definition of different formulas for the weights and the empirical evaluation of the measures defined as stopping rule criteria.
Optimal experimental design from observation data: weights and stopping rule / R. BERNI. - In: STATISTICA APPLICATA. - ISSN 1125-1964. - STAMPA. - 17:(2005), pp. 5-24.
Optimal experimental design from observation data: weights and stopping rule
BERNI, ROSSELLA
2005
Abstract
In this paper we focus on the use of large data sets to implement optimal designs. Our final goal is an experimental design that is optimal by means of sequential designs. In this work the improvement is related to two aspects: weights and stopping rule. The definition of weights is relevant to discriminate between more than two models. The settlement of a stopping rule is important to evaluate the contribution of the new selected experimental point. The main problems here faced are the choice of the best model according to the definition of different formulas for the weights and the empirical evaluation of the measures defined as stopping rule criteria.I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.