Logrank

Compare the survival distributions of two samples by performing Log-rank test

Description

Log-rank test(Approximate)
The logrank test is a hypothesis test to compare the survival distributions of two samples. It is a nonparametric test and appropriate to use when the data are right censored (technically, the censoring must be non-informative).
logrank(x,y,2)
return p value and graph K-M curves

• Required Products: Statistics Toolbox
• MATLAB release: MATLAB 7.4 (R2007a)

[Jos x@y.z]

Why zip a single m-file? And there is no penalty for using spacing and indents, which making the text so much more readable.
The help section is a little vague mentioning only two arguments for its syntax and then talk about a three more variables : "Syntax: p=logrank(x,y) Input: x and y - two groups; show -0:don't show the result ..."

And maybe you can provide an example what this test is all about: when to use it for what, and what are its advantages over other tests.

[Fan Lin]

Thanks for your suggestions. I will improve it next time.

Logrank test is a classic method to compare two populations with censored survival data
and there is an example in the The help section

[Fan Lin]

John D'Errico?
Thankyou,I had updated this file.