K-sample tests based on survival analysis with interval-censored data

• Main Authors: Elfaki, Faiz Ahmed Mohamed, Daud, Isa, Noor Akma, Ibrahim, Jamal Ibrahim, Daoud, Usman, Mustofa, Lukman, Iing
• Format: Article
• Language: English
• Published: ILRAM Publisher 2009
• Subjects: QA Mathematics
• Online Access: http://irep.iium.edu.my/6271/1/K-sample_tetsts_based_on_survival_analysis_with_interval-_-censored_data.pdf; http://irep.iium.edu.my/6271/; http://www.ijset.org/index.php/ijset/index
• Institution: Universiti Islam Antarabangsa Malaysia

The nonparametric tests are introduced and provides a new method to compare survival functions among groups when the data are interval-censored. The k-sample test for interval censored is studied. Many k-sample tests have been developed when data are uncensored or right censored. The main characteristic of this paper is the permutational distributions of survival analysis with interval-censored data The difficulty of finding the distribution of the test statistic is avoided with this permutational approach. The permutational methodology will be introduced and provides description of the permutational Wilcoxon and Log-rank tests for interval-censored data. A likelihood approach for this new proposal is also considered. The relationship given by Fay (1999) betWeen permutational methods and likelihood methods based on score test statistics will be extended. It was found that the permutational approach requires that the censoring mechanisms are identical across the groups.