04/10/2022 - 05/10/2022
Royal Statistical Society
Dr Dave Collett
Advanced (specialised prior knowledge)
View in Google Maps (EC1Y 8LX)
12 Errol Street, London
Level: Professional (P)
Standard methods of survival analysis based on the Kaplan-Meier estimate of a survivor function, the log rank test and Cox regression modelling are widely used in many different areas of application. But often, the assumptions that underlie these techniques may not be valid, or the data structure may be more complex. Extensions of these basic methods allow particular features of data that occur in practice to be handled appropriately. This course will begin with an overview of standard methods and then move on to some of the more advanced techniques. Their practical application will be illustrated using SAS and R.
An appreciation of how the methods of survival analysis can be used in a variety of situations.
Overview of standard methods for summarising survival data and the Cox regression model. Types of censoring in survival data, including interval and dependent censoring. Time dependent variables and the counting process formulation of survival data. Parametric models for survival data, including flexible models based on splines. Incorporating random effects into a survival analysis; frailty models. Analysis of data where there is more than one type of event; models for competing risks. Detecting and handling non proportional hazards.
Statisticians and epidemiologists in public sector research organisations, pharmaceutical companies and related organisations. University research students and fellows.
Some familiarity with basic methods for summarising survival data, including estimates of the survivor function and the log rank test. Some experience in using the Cox regression model would be advantageous.
£599.76 to £832.32 (including VAT)
Website and registration:
Quantitative Data Handling and Data Analysis, Survival analysis , Cox regression model , Survival data , Censoring , Frailty models
Related publications and presentations: