Bayesian Latent Class Models in malaria diagnosis

The main focus of this study is to illustrate the importance of the statistical analysis in the evaluation of the accuracy of malaria diagnostic tests, without admitting a reference test, exploring a dataset (n=3317) collected in São Tomé and Príncipe.
Read More

Conditional dependence diagnostic in the latent class model: A simulation study

Authors: de Oliveira MR, Gonçalves L, Subtil A
Journal: Statistics & Probability Letters
Link: https://apps.webofknowledge.com/full_record.do?product=UA&search_mode=GeneralSearch&qid=4&SID=T2VvZFAKYvQv3Eo9tzK&page=1&doc=7

The classical latent class model assumes the hypothesis of conditional independence. We explore tools commonly used to validate this hypothesis (correlation residual plot, log-odds ratio check plot, and known goodness of fit tests) to make practitioners aware of these tools’ shortcomings in correctly identifying local dependence.
Read More

Bayesian Latent Class Models in malaria diagnosis

Conditional dependence diagnostic in the latent class model: A simulation study

About GHTM

Contacts

Map