Autori
Sankaran, Paduthol Godan
Rejani, Parassery Parameswaran

Titolo
Regression Analysis of Cure Model with Generalised Weibull Distribution
Periodico
Statistica
Anno: 2021 - Volume: 81 - Fascicolo: 3 - Pagina iniziale: 265 - Pagina finale: 278

Cure models are of special attention when all of the study subjects do not experience the event of interest even after long follow-up time. Many researchers have used exponential, gamma and Weibull distribution in the latency part of parametric cure models. In this article, we propose a new regression model with cured fraction, in its latency part is explained by the generalised Weibull distribution (Mudholkar et al., 1996). The estimation of the parameters of the proposed model is done using maximum likelihood method via EM algorithm. Simulations are carried out to study the effect of sampling fluctuations and to knowthe efficiency of estimators. The proposed model is applied to real data on acute myelogenous leukaemia. The statistical significance of the regression parameter is checked by likelihood ratio (LR) test and the new model was compared withWeibull cure model using Akaike information criterion (AIC).



SICI: 0390-590X(2021)81:3<265:RAOCMW>2.0.ZU;2-S

Esportazione dati in Refworks (solo per utenti abilitati)

Biblioteche ACNP che possiedono il periodico