Cutpoint determination methods in competing risks subdistribution model
In the analysis involving clinical and psychological data, by transforming a continuous predictor variable into a categorical variable, usually binary, a more interpretable model can be established. Thus, we consider the problem of obtaining a threshold value of a continuous covariate given a com...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Penerbit ukm
2009
|
| Online Access: | http://journalarticle.ukm.my/1926/ http://www.ukm.my/~ppsmfst/jqma/index.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In the analysis involving clinical and psychological data, by transforming a continuous
predictor variable into a categorical variable, usually binary, a more interpretable model can
be established. Thus, we consider the problem of obtaining a threshold value of a continuous
covariate given a competing risk survival time response using a binary partitioning algorithm
as a way to optimally partition data into two disjoint sets. Five cutpoint determination methods
are developed based on regression of competing risks subdistribution. Simulation results show
that the deviance method has the desired properties. Permutation test is used to assess the level
of significance and bootstrap confidence interval is obtained for the optimal cutpoint. The
deviance method is applied to determine cutpoint of age for a real dataset |
|---|