quantile.survfit.RdRetrieve quantiles and confidence intervals for them from a survfit or Surv object.
# S3 method for class 'survfit'
quantile(x, probs = c(0.25, 0.5, 0.75), conf.int = TRUE,
scale, tolerance= sqrt(.Machine$double.eps), ...)
# S3 method for class 'survfitms'
quantile(x, probs = c(0.25, 0.5, 0.75), conf.int = TRUE,
scale, tolerance= sqrt(.Machine$double.eps), ...)
# S3 method for class 'survfit'
median(x, ...)a result of the survfit function
numeric vector of probabilities with values in [0,1]
should lower and upper confidence limits be returned?
optional scale factor, e.g., scale=365.25 would
return results in years if the fit object were in days.
tolerance for checking that the survival curve exactly equals one of the quantiles
optional arguments for other methods
The kth quantile for a survival curve S(t) is the location at which a horizontal line at height p= 1-k intersects the plot of S(t). Since S(t) is a step function, it is possible for the curve to have a horizontal segment at exactly 1-k, in which case the midpoint of the horizontal segment is returned. This mirrors the standard behavior of the median when data is uncensored. If the survival curve does not fall to 1-k, then that quantile is undefined.
In order to be consistent with other quantile functions, the argument
prob of this function applies to the cumulative distribution
function F(t) = 1-S(t).
Confidence limits for the values are based on the intersection of the
horizontal line at 1-k with the upper and lower limits for the
survival curve. Hence confidence limits use the same
p-value as was in effect when the curve was created, and will differ
depending on the conf.type option of survfit.
If the survival curves have no confidence bands, confidence limits for
the quantiles are not available.
When a horizontal segment of the survival curve exactly matches one of
the requested quantiles the returned value will be the midpoint of the
horizontal segment; this agrees with the usual definition of a median
for uncensored data. Since the survival curve is computed as a series
of products, however, there may be round off error.
Assume for instance a sample of size 20 with no tied times and no
censoring. The survival curve after the 10th death is
(19/20)(18/19)(17/18) ... (10/11) = 10/20, but the computed result will
not be exactly 0.5. Any horizontal segment whose absolute difference
with a requested percentile is less than tolerance is
considered to be an exact match.
The quantiles will be a vector if the survfit object contains
only a single curve, otherwise it will be a matrix or array. In
this case the last dimension will index the quantiles.
If confidence limits are requested, then result will be a list with
components
quantile, lower, and upper, otherwise it is the
vector or matrix of quantiles.
fit <- survfit(Surv(time, status) ~ ph.ecog, data=lung)
quantile(fit)
#> $quantile
#> 25 50 75
#> ph.ecog=0 285 394 655
#> ph.ecog=1 181 306 550
#> ph.ecog=2 105 199 351
#> ph.ecog=3 118 118 118
#>
#> $lower
#> 25 50 75
#> ph.ecog=0 189 348 558
#> ph.ecog=1 156 268 460
#> ph.ecog=2 61 156 285
#> ph.ecog=3 NA NA NA
#>
#> $upper
#> 25 50 75
#> ph.ecog=0 350 574 NA
#> ph.ecog=1 223 429 689
#> ph.ecog=2 163 288 654
#> ph.ecog=3 NA NA NA
#>
cfit <- coxph(Surv(time, status) ~ age + strata(ph.ecog), data=lung)
csurv<- survfit(cfit, newdata=data.frame(age=c(40, 60, 80)),
conf.type ="none")
temp <- quantile(csurv, 1:5/10)
temp[2,3,] # quantiles for second level of ph.ecog, age=80
#> 10 20 30 40 50
#> 92 144 181 218 270
quantile(csurv[2,3], 1:5/10) # quantiles of a single curve, same result
#> 10 20 30 40 50
#> 92 144 181 218 270