estMargProb {discSurv} | R Documentation |
Estimates the marginal probability P(T=t|x) based on estimated discrete hazards. The discrete hazards may or may not depend on covariates. The covariates have to be equal across all estimated hazard rates. Therefore the given discrete hazards should only vary over time.
estMargProb(hazards)
hazards |
Estimated discrete hazards ("numeric vector") |
The argument hazards must be given for all intervals [a_0, a_1), [a_1, a_2), ..., [a_q-1, a_q), [a_q, Inf).
Estimated marginal probabilities ("numeric vector")
It is assumed that all time points up to the last interval [a_q, Inf) are available. If not already present, these can be added manually.
Thomas Welchowski welchow@imbie.meb.uni-bonn.de
Tutz G, Schmid M (2016). Modeling discrete time-to-event data. Springer Series in Statistics.
# Example unemployment data
library(Ecdat)
data(UnempDur)
# Select subsample
subUnempDur <- UnempDur [1:100, ]
# Convert to long format
UnempLong <- dataLong(dataShort = subUnempDur, timeColumn = "spell", eventColumn = "censor1")
head(UnempLong)
# Estimate binomial model with logit link
Fit <- glm(formula = y ~ timeInt + age + logwage, data = UnempLong, family = binomial())
# Estimate discrete survival function given age, logwage of first person
hazard <- predict(Fit, newdata = subset(UnempLong, obj == 1), type = "response")
# Estimate marginal probabilities given age, logwage of first person
MarginalProbCondX <- estMargProb (c(hazard, 1))
MarginalProbCondX
sum(MarginalProbCondX)==1 # TRUE: Marginal probabilities must sum to 1!