Loss Due to Missing Data in Efficiency of a Locally Optimal Test for Homogeneity with Respect to Very Rare Events*
AUTOR(ES)
Puri, Prem S.
RESUMO
With reference to observations of supernovae in galaxies, a locally optimal test for the hypothesis of homogeneity of the observational units with respect to occurrence of supernovae (treated as a very rare event) is obtained for the case where the data are available only for those galaxies where at least one supernova is observed. It is shown that the loss in the asymptotic efficiency of the test due to lack of reporting of galaxies with no supernovae is very heavy and in fact is infinite for the case of very rare events.
ACESSO AO ARTIGO
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=283269Documentos Relacionados
- LOCALLY OPTIMAL TEST FOR HOMOGENEITY WITH RESPECT TO VERY RARE EVENTS*
- Evaluation of the rapid diagnostic test OptiMAL for diagnosis of malaria due to Plasmodium vivax
- Efficiency loss in HPGe detectors due to beta and gamma sum coincidence
- Comparison of the OptiMAL Test with PCR for Diagnosis of Malaria in Immigrants
- Loss to follow-up and missing data: important issues that can affect your study results