In this paper we perform new Monte Carlo simulation studies to investigate the sample behaviour of estimates in a binary response multilevel model. Particularly, we analyse the influence of different factors (small sample sizes, unbalanced and balanced groups and a model specification with both continuous and dummy variables) on the accuracy of estimates and standard errors of estimates. In all instances we have considered the point estimates are unbiased (even with very small sample sizes), whereas variance components are underestimated. The accuracy of the standard errors of variance estimates needs a very large number of groups.
The Accuracy of Estimates in Discrete Response Multilevel Models: New Simulation Results
PACCAGNELLA, OMAR
2003
Abstract
In this paper we perform new Monte Carlo simulation studies to investigate the sample behaviour of estimates in a binary response multilevel model. Particularly, we analyse the influence of different factors (small sample sizes, unbalanced and balanced groups and a model specification with both continuous and dummy variables) on the accuracy of estimates and standard errors of estimates. In all instances we have considered the point estimates are unbiased (even with very small sample sizes), whereas variance components are underestimated. The accuracy of the standard errors of variance estimates needs a very large number of groups.File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
