Discrepancy in logrank p-value from `analyse_survival` for ≥3 level factors

cran / survivalAnalysis

:exclamation: This is a read-only mirror of the CRAN R package repository. survivalAnalysis — High-Level Interface for Survival Analysis and Associated Plots

3 stars 1 forks source link

Discrepancy in logrank p-value from `analyse_survival` for ≥3 level factors #1

Open KMBHPB opened 4 years ago

KMBHPB commented 4 years ago

Hi,

Firstly, great package - lots of handy features and easy to generate the required information.

However, when I apply analyse_survival for UV analysis to any factors with ≥3 levels, the overall analysis p-value is consistently smaller than that returned by survdiff from survival.

The same is not observed for any two level factors.

Could you perhaps explain this discrepancy??

Cheers,

Kai

gaborcsardi commented 4 years ago

Hi, this is a read only mirror of CRAN, please see the package authors in the DESCRIPTION file. Thanks!