FREQUENTIST AND BAYESIAN APPROACH FOR THE MIXTURE CURE MODELS WITH GENERALIZED POWER GENERALIZED WEIBULL BASELINE: AN APPLICATION TO CANCER DATA
This paper develops the Weibull distribution by adding a new parameter to the classical distribution, the generalized power generalized Weibull mixture cure model distribution, and it’s extremely useful when modeling survival data with parameter hazard rate function shape. As a result, there is greater flexibility in analyzing and modeling various data types. The essential mathematical and statistical characteristics of the proposed distribution are generated. In this paper. Many well-known life time special sub-models, such as Rayleigh, Power generalized Weibull, Nadarajah-Haghighi, Weibull, and Exponential, are included in the proposed distribution. The maximum likelihood distribution method was used to estimate the unknown parameters of the proposed distribution; and the effectiveness of the estimators was determined using Markov Chain Monte Carlo simulation study. The Markov Chain Monte Carlo used to develop diagnostic methods. This distribution is important because it can model non-monotone and monotone, upside-down bathtub, and bathtub hazard rate functions. All of which are widely used in survival and efficiency data analysis. Moreover, the flexibility and effectiveness of the proposed distribution are demonstrated in a real-world data set and compared to its sub models. Based on the goodness of fit and in- formation criterion value, the proposed distribution is accurate. Finally, the estimation of the data set is determined using Bayesian inference and Gibb’s sampling performance. In addition to Bayes estimates, the highest posterior density reliable intervals and Markov Chain Monte Carlo convergence diagnostic technique were used.
Bathtub, Bayesian, Classical Approach, Hazard Rate, Maximum Likelihood Estimation, Monotonic and Non-Monotonic.