| dsurvreg {survival} | R Documentation |
Density, cumulative distribution function, quantile function and random
generation for the set of distributions
supported by the survreg function.
dsurvreg(x, mean, scale=1, distribution='weibull', parms) psurvreg(q, mean, scale=1, distribution='weibull', parms) qsurvreg(p, mean, scale=1, distribution='weibull', parms) rsurvreg(n, mean, scale=1, distribution='weibull', parms)
x |
vector of quantiles.
Missing values ( |
q |
vector of quantiles.
Missing values ( |
p |
vector of probabilities.
Missing values ( |
n |
number of random deviates to produce |
mean |
vector of linear predictors for the model.
This is replicated to be the same length as |
scale |
vector of (positive) scale factors.
This is replicated to be the same length as |
distribution |
character string giving the name of the distribution. This must be one
of the elements of |
parms |
optional parameters, if any, of the distribution. For the t-distribution this is the degrees of freedom. |
Elements of q or
p that are missing will cause the corresponding
elements of the result to be missing.
The location and scale
values are as they would be for survreg.
The label "mean" was an unfortunate choice (made in mimicry of qnorm);
since almost none of these distributions are symmetric it will not
actually be a mean, but corresponds instead to the linear predictor of
a fitted model.
Translation to the usual parameterization found in a textbook is not
always obvious.
For example, the Weibull distribution is fit using the
Extreme value distribution along with a log transformation.
Letting F(t) = 1 - exp(-(at)^p)
be the cumulative distribution of the
Weibull using a standard parameterization in terms of
a and p,
the survreg location corresponds to -log(a) and the scale
to 1/p
(Kalbfleisch and Prentice, section 2.2.2).
density (dsurvreg),
probability (psurvreg),
quantile (qsurvreg), or
for the requested distribution with mean and scale
parameters mean and
sd.
Kalbfleisch, J. D. and Prentice, R. L. (1970). The Statistical Analysis of Failure Time Data Wiley, New York.
# List of distributions available names(survreg.distributions) ## Not run: [1] "extreme" "logistic" "gaussian" "weibull" "exponential" [6] "rayleigh" "loggaussian" "lognormal" "loglogistic" "t" ## End(Not run) # Compare results all.equal(dsurvreg(1:10, 2, 5, dist='lognormal'), dlnorm(1:10, 2, 5)) # Hazard function for a Weibull distribution x <- seq(.1, 3, length=30) haz <- dsurvreg(x, 2, 3)/ (1-psurvreg(x, 2, 3)) ## Not run: plot(x, haz, log='xy', ylab="Hazard") #line with slope (1/scale -1) ## End(Not run)