Analysis of Clustered Women Birth Interval in Malawi: Application of Parametric and Non Parametric Survival Mixed-Effects Regression under Various Distributions of Random Effects and Priors.

Chikumbutso Wasela, Unima

The analysis of birth intervals has emerged as a key factor in determining population fertility rates in recent years. Several observational studies have shown the relationship between the determinate and the birth interval. These factors might not be able to completely explain the actual variations in birth interval. This could be the result of unobserved variable that the model ignores. Including frailty in the model can correctly measure the effects of the covariates on the birth interval. The Cox model is extended to parametric, semi-parametric, and Bayesian non-parametric frailty models in this work. Using women's clustered data from the Malawi Demographic Survey 2015–2016, we examine the effectiveness of several frailty distribution models. R 4.1.2 is used for all data analysis. While DIC is utilized in Bayesian non-parametric models, AIC is used to compare the performance of various frailty model types in semi-parametric and parametric models.