presmoothTP: This function implements presmoothed estimates based on the...

In presmoothedTP: Presmoothed Landmark Aalen-Johansen Estimator of Transition Probabilities for Complex Multi-State Models

This function implements presmoothed estimates based on the landmark Aalen-Johansen method for non-parametric estimation of transition probabilities in non-Markov multi-state models.

Description

This function is used to obtain nonparametric presmoothed estimates of the transition probabilities for any type of multistate model, including models that allow for reversible transitions, and potencially non-Markovian seetings. The methods used to compute the transition probabilities are ”PLM” (given by a ”logistic” model), ”cauchit”, ”probit”. The proposed presmoothed estimator is designed to handle complex multi-state models with more than three states, as well as reversible transitions, accommodating potentially non-Markovian behaviors.

Usage

presmoothTP(data, s, from, to, method = "LM", trans)

Arguments

data	A data frame in the long format containing the subject id; from corresponding to the starting state; the receiving state, to; the transition number, trans; the starting time of the transition given by Tstart; the stopping time of the transition, Tstop, and status for the status variable, with 1 indicating an event (transition), 0 a censoring.
s	The first time for obtaining estimates for the transition.
from	The starting state of the transition probabilities.
to	The last receiving state considered for the estimation of the transition probabilities.
method	The method used to compute the transition probabilities. Possible options are "PLM", "cauchit" and "probit".
trans	The transition matrix for multi-state model.

Value

Nonparametric presmoothed estimates of transition probabilities based on logistic’, ’probit’, and ’cauchit’ distributions (PLM,cauchit and probit).

Author(s)

Gustavo Soutinho and Luis Meira-Machado.

Examples

data("ebmt4")
db_wide <- ebmt4
positions<-list(c(2, 3, 5, 6), c(4, 5, 6), c(4, 5, 6), c(5, 6), c(6), c())
states.names = c("Tx", "Rec", "AE", "Rec+AE", "Rel", "Death")
trans<-transTP(positions, states.names)
times.names = c(NA, "rec", "ae","recae", "rel", "srv")
status.names=c(NA, "rec.s", "ae.s", "recae.s", "rel.s", "srv.s")
data.long<-prepTP(data=db_wide, trans, times.names, status.names)

res1<-presmoothTP(data.long, s = 100, from = 1, to = 1, 
method = 'PLM', trans=trans)
res2<-presmoothTP(data.long, s = 100, from = 1, to = 1, method = 'cauchit', 
trans=trans)
res3<-presmoothTP(data.long, s = 100, from = 1, to = 1, method = 'probit', 
trans=trans)

plot(res1, type='s', xlim=c(100,3000), ylim=c(0.6,1), ylab=c('p11(100,t)'))
lines(res2, type='s', col=2)
lines(res3, type='s', col=4)
legend("topright", legend = c("PLM","Cauchit", "Probit"), lwd = 0.3, 
col = c(1,2,4), cex=0.7, lty=c(1, 1))
res4<-presmoothTP(data.long, s = 1000, from = 2, to = 6, method = 'PLM', 
trans=trans)
res5<-presmoothTP(data.long, s = 1000, from = 2, to = 6, method = 'cauchit', 
trans=trans)
res6<-presmoothTP(data.long, s = 1000, from = 2, to = 6, method = 'probit', 
trans=trans)
plot(res4, type='s', xlim=c(1000,3000), ylim=c(0,0.12),ylab=c('p26(1000,t)'))
lines(res5, type='s', col=2)
lines(res6, type='s', col=4)
legend("topright", legend = c("PLM","Cauchit", "Probit"), lwd = 0.3, 
col = c(1,2,4), cex=0.7, lty=c(1, 1))

presmoothedTP documentation built on April 3, 2025, 7:44 p.m.