An Alternative Promotion Time Cure Model with Overdispersed Number of Competing Causes: An Application to Melanoma Data

A cure rate model under the competing risks setup is proposed. For number of causes related to occurrence event interest, we posit one-parameter Bell distribution, which accommodates overdispersed counts. The parameterized in rate, linked covariates. Parameter estimation based on maximum likelihood method. Estimates are computed via EM algorithm. In order compare different models, a selection criterion for non-nested models implemented. Results from simulation studies indicate that method and have good performance. dataset melanoma analyzed using proposed as well some literature.