The Log-rank test

Source: R/logrank.R

Performs the log-rank test on survival data, possibly stratified.

logrank(Y, group, data = parent.frame())

Arguments

Y

a survival object as returned by the Surv function.
group

defines the groups to be compared. Coerced to a factor.
data

a data.frame in which to interpret the variables.

Value

A list of class logrank with components

test.statistic

The logrank (score) test statistic.

df

The degrees of freedom of the test statistic.

p.value

The p value of the test.

hazards

A list of two-column matrices, describing event times and corresponding hazard atoms in each stratum (class 'hazdata').

call

The call

Note

The test is performed by fitting a Cox regression model and reporting its score test. With tied data, this might be slightly different from the true logrank test, but the difference is unimportant in practice.

See also

coxreg, print.logrank.

Author

Göran Broström

Examples

fit <- logrank(Y = Surv(enter, exit, event), group = civ, data = oldmort)
fit

#> 
#>      The log-rank test
#> 
#> Call:
#> logrank(Y = Surv(enter, exit, event), group = civ, data = oldmort)
#> 
#>  X-squared =  16.2888 , df =  2 , p-value =  0.000290358