 |
 |
Dear all, 1) The Centre for Multilevel Modelling's REALCOM-Impute software has been updated. This version fixes a bug whereby when there area large number of responses machine accuracy can result in infinite estimates for parameters and hence imputed values. There is still an issue we are studying when binary responses are declared as ordered. So, for now REALCOM-Impute will treat all binary responses as unordered. Since this is mathematically equivalent this should cause no problems. Further details and REALCOM-Impute download: http://www.bristol.ac.uk/cmm/software/realcom/imputation.html 2) Leonardo Grilli and Carla Rampichini, Professors of Statistics at the University of Florence, have published extensively on methods for causal inference, multilevel modelling and duration modelling, with applications in educational research. The presentations they recently gave here at the University of Bristol are now available: http://www.bristol.ac.uk/cmm/software/mlwin/support/workshops/materials/causal-inference/ Many other previous workshop presentations can be downloaded from our web site: http://www.bristol.ac.uk/cmm/software/support/workshops/materials/ With best regards, Hilary Browne Centre for Multilevel Modelling University of Bristol 2 Priory Road Bristol BS8 1TX UK You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.