Survival analysis plays a crucial role in medical research for understanding the time until an event of interest occurs, such as disease recurrence or death. An important branch of survival analysis models is cure models, assuming that a proportion of subjects will never experience the event of interest. The value of the proportion is called the cured rate and is usually associated with many covariates with complex effect relationships. Studying cure models under such non-linear covariate effects remains an active research area. This thesis aims to investigate advancements in additive cure models, focusing on their ability to capture additive complex relationships between covariates and survival outcomes with a cured
fraction through non-linear modeling techniques, such as basic splines. Additive cure models offer a robust framework for analyzing survival data when a subset of individuals is cured and does not experience the event. The thesis will involve simulation studies to assess the accuracy of parameter estimation and model fit in various scenarios, and the application of additive cure models to real-world datasets from medical research studies. The findings will enhance the understanding and application of additive cure models in analyzing survival
data with non-linear covariate effects, with implications for clinical decision-making and prognostic modeling. The insights gained from this research have implications for various fields, including epidemiology, clinical research, and public health, providing valuable tools for analyzing survival data and enhancing decision-making processes.