presmTP {presmTP} | R Documentation |
Methods for estimation of transition probabilities in the illness-death model
Description
This function is used to obtain unsmoothed and presmoothed estimates of the transition probabilities in the illness-death model.
Usage
presmTP(data, s, method = "uns", estimand = "S",
bw.selec = "plug-in", fixed.bw = NULL, bound = "none")
Arguments
data |
A numeric value to be squared |
s |
The first time for obtaining estimates for the transition probabilities. |
method |
The method used to compute the transition probabilities.
Possible options are |
estimand |
An optional character string identifying the function to estimate: "S" for survival function and "H" for cumulative hazard function. Defaults to "S". |
bw.selec |
An optional (partially matched) character string specifying the method of bandwidth selection. "fixed" if no bandwidth selection is done, in which case the bandwidth(s) given by the fixed.bw argument is (are) used, "plug-in" for plug-in bandwidth selection and "bootstrap" for bootstrap bandwidth selection. Defaults to "fixed". |
fixed.bw |
An optional numeric vector with the fixed bandwidth(s) used when the value of the bw.selec argument is "fixed". It must be of length 1 for estimating survival and cumulative hazard functions, and of length 2 for density and hazard functions (in this case, the first element is the presmoothing bandwidth). |
bound |
An optional numeric vector with the fixed bandwidth(s) used when the value of the bw.selec argument is "fixed". It must be of length 1 for estimating survival and cumulative hazard functions, and of length 2 for density and hazard functions (in this case, the first element is the presmoothing bandwidth). |
Value
An object of class "pstp" and one of the following classes: "uns"
,
"np"
, "logit"
, "logit.gam"
,
"probit"
and "cauchit"
. Objects are implemented as a list with elements:
est0 |
data.frame with estimates of the transition probabilities 0->0, 0->1 and 0->2. |
est1 |
data.frame with estimates of the transition probabilities 1->1 and 1->2. |
s |
The first time for obtaining estimates for the transition probabilities. |
callp |
The expression of the estimated probability. |
call |
A call object. |
Author(s)
Gustavo Soutinho, Luis Meira-Machado, Pedro Oliveira.
References
Aalen O. O., Johansen S. (1978) An Empirical Transition Matrix for Nonhomogeneous Markov Chains Based on Censored Observations. Scandinavian Journal of Statistics 5(3), 141–150.
Cao, R., Lopez-de-Ullibarri, I., Janssen, P. and Veraverbeke, N. (2005). Presmoothed Kaplan-Meier and Nelson-Aalen estimators, Journal of Nonparametric Statistics, 17, 31-56.
Meira-Machado L. F., de Una-Alvarez J. and Cadarso-Suarez C. (2006). Nonparametric estimation of transition probabilities in a non-Markov illness-death model. Lifetime Data Anal 12(3), 325–344.
Lopez-de-Ullibarri, I and Jacome, M. A. (2013). survPresmooth: An R Package for Presmoothed Estimation in Survival Analysis, Journal of Statistical Software, 54(11), 1-26. URL: http://www.jstatsoft.org/v54/i11/. de Una-Alvarez J. and Meira-Machado L. (2015). Nonparametric estimation of transition probabilities in a non-Markov illness-death model: a comparative study. Biometrics 71, 364–375.
Meira-Machado, L. (2016). Smoothed landmark estimators of the transition probabilities, SORT-Statistics and Operations Research Transactions, 40, 375-398.
Examples
#Unsmoothed
res1<- presmTP(data = colonIDM, s = 365,method = "uns" )
res1$est0$t
res1$est0$p02
res1$est1$t
summary(res1, state_ini=1, time=365*1:5)
plot(res1)
res1$call
class(res1)
#Nonparametric
res2<- presmTP(data = colonIDM, s = 365,method = "np" )
res3<- presmTP(data = colonIDM, s = 365,method = "np", estimand="S")
res4<- presmTP(data = colonIDM, s = 365,method = "np", estimand="H")
res5<- presmTP(data = colonIDM, s = 365,method = "np",
bw.selec="fixed", fixed.bw=30)
#Presmoothed - Logit
res6<- presmTP(data = colonIDM, s = 365,method = "logit" )
summary(res6, state_ini=1, time=365*1:5)
#Presmoothed - Logit GAM
res7<- presmTP(data = colonIDM, s = 365,method = "logit.gam" )