Student Seminar Series - August 24, 2006
University of Minnesota
School of Statistics
College of Liberal Arts
A Comparison
Analysis
between Cox’s Proportional Hazards Model and Aalen’s Additive Hazards
Regression Model Using Veteran’s Administration Lung Cancer Data
Wen Ouyang
Thursday, August 24, 2006
10:00 AM, B60
Ford Hall
Minneapolis, East Bank Campus
Refreshments at 9:30 AM
300 Ford Hall
Abstract
Cox’s
proportional hazards model is one of the most well-known and widely
used models
in survival analysis. Relative risks are easily calculated and easy to
interpret. However, it is difficult to
model covariate effects over time. Aalen’s
additive hazards regression model assumes that covariates act in an
additive
manner on an unknown baseline hazard rate and the effect of a covariate
may
vary over time. Hence, Aalen’s
model may provide more in depth information on the effect of a
prognostic
factor over time. However, Aalen’s
model is not available in some commonly used computer packages, such as
SAS or
SPSS. In this project we use R “addreg” package to perform Aalen’s model.
The aim of this study is to
compare two models in terms of theory and application to a real data
set which
is from Veteran's Administration Lung Cancer Trial. Two models have
similar
results with regard to covariates selected for the model. In
particular, both
models suggest that treatment type has no significant effect. However,
two
models give different information and should not be viewed as
alternatives to
each other, but as complementary methods that may be used together to
give a more
complete and more comprehensive understanding of data.