Can neural networks help statistics in hard conditions? The present article discusses the usefulness of the simulation, with neural networks in particular, to develop strategies for statistical testing of hypotheses. Starting from a database regarding mental imagination in groups differentiated by age, level of education, presence and degree of cognitive impairment, traditional statistical analyses have been compared with those derived from models based on neural networks, able to learn a general model from which large numbers of simulated replications can be deduced, without altering the internal composition of the samples, thus increasing the power of the multivariate analysis with respect to the original database. Some hypotheses are proposed for exploiting neural networks to analyze data in complex applied researches.

Le reti neurali possono aiutare la statistica in situazioni ‘difficili’?

DI CORRADO, DONATELLA;GUARNERA, MARIA ANTONELLA ELISABETTA;DI NUOVO, ALESSANDRO
2015

Abstract

Can neural networks help statistics in hard conditions? The present article discusses the usefulness of the simulation, with neural networks in particular, to develop strategies for statistical testing of hypotheses. Starting from a database regarding mental imagination in groups differentiated by age, level of education, presence and degree of cognitive impairment, traditional statistical analyses have been compared with those derived from models based on neural networks, able to learn a general model from which large numbers of simulated replications can be deduced, without altering the internal composition of the samples, thus increasing the power of the multivariate analysis with respect to the original database. Some hypotheses are proposed for exploiting neural networks to analyze data in complex applied researches.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11387/108132
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact