Lectures and seminars CBB biostatistics seminar with Therese Andersson: "Flexible Parametric Survival Models as an alternative to Cox regression"
Centre for bioinformatics and biostatistics presents a biostatistics seminar.
The Cox model is a commonly used regression model for survival data. One of the advantages of the Cox model is that the baseline hazard is not estimated, and therefore, one doesn’t have to make any assumptions about the shape of the underlying baseline hazard function. The disadvantage is that the underlying shape of the hazard is then often ignored, even though it could be of interest. Also, it can easier to model violations to the proportional hazards assumption from models where the baseline hazard is estimated. In this presentation, Dr ThereseAndersson will describe the flexible parametric survival model as an alternative to Cox regression. The flexible parametric model use splines to capture the shape of the underlying hazard function, which enables easy estimation of hazard rates together with hazard ratios. Non-proportional hazards can be allowed by including interactions with the splines for time.
Before the seminar, a limited number of sandwiches and drinks will be served on a “First come – first served” basis starting 12:00. Show up early and network!
The seminar is held in English.