Provides estimates for the transition probabilities based on presmoothed Kaplan-Meier weighted estimators, KMPW.
transKMPW(object, s, t, state.names=c("1", "2", "3"), conf=FALSE, n.boot=1000,
conf.level=0.95, method.boot="percentile", method.est=3)An object of class ‘TPmsm’. There are methods for contour, image, print and plot.
‘TPmsm’ objects are implemented as a list with elements:
A string indicating the type of estimator used in the computation.
A matrix with transition probability estimates. The rows being the event times and the columns the 5 possible transitions.
A matrix with the lower transition probabilities of the confidence band. The rows being the event times and the columns the 5 possible transitions.
A matrix with the upper transition probabilities of the confidence band. The rows being the event times and the columns the 5 possible transitions.
Vector of times where the transition probabilities are computed.
Start of the time interval.
End of the time interval.
The bandwidth used. If the estimator doesn't require a bandwidth, it's set to NULL.
A vector of characters giving the states names.
Number of bootstrap samples used in the computation of the confidence band.
Level of confidence used to compute the confidence band.
An object of class ‘survTP’.
The first time for obtaining estimates for the transition probabilities. If missing, 0 will be used.
The second time for obtaining estimates for the transition probabilities.
If missing, the maximum of Stime will be used.
A vector of characters giving the state names.
Provides pointwise confidence bands. Defaults to FALSE.
The number of bootstrap samples. Defaults to 1000 samples.
Level of confidence. Defaults to 0.95 (corresponding to 95%).
The method used to compute bootstrap confidence bands. Possible options are “percentile” and “basic”. Defaults to “percentile”.
The method used to compute the estimate. Possible options are 1, 2, 3 or 4.
Artur Araújo, Javier Roca-Pardiñas and Luís Meira-Machado
If method.est=1 then \(p_{11}(s,t)\), \(p_{12}(s,t)\) and \(p_{22}(s,t)\) are estimated according to the following expressions:
\(p_{11}(s,t)=\frac{1-P(Z \leq t)}{1-P(Z \leq s)}\),
\(p_{12}(s,t)=\frac{P(Z \leq t)-P(Z \leq s)-P(s<Z \leq t, T \leq t)}{1-P(Z \leq s)}\),
\(p_{22}(s,t) =\frac{P(Z \leq s)-P(Z \leq s,T \leq t)}{P(Z \leq s)-P(T \leq s)}\).
Then, \(p_{13}(s,t)=1-p_{11}(s,t)-p_{12}(s,t)\) and \(p_{23}(s,t)=1-p_{22}(s,t)\).
If method.est=2 then \(p_{11}(s,t)\), \(p_{12}(s,t)\) and \(p_{22}(s,t)\) are estimated according to the following expressions:
\(p_{11}(s,t)=\frac{P(Z>t)}{P(Z>s)}\),
\(p_{12}(s,t)=\frac{P(s<Z \leq t,T>t)}{P(Z>s)}\),
\(p_{22}(s,t) =\frac{P(Z \leq s,T>t)}{P(Z \leq s,T>s)}\).
Then, \(p_{13}(s,t)=1-p_{11}(s,t)-p_{12}(s,t)\) and \(p_{23}(s,t)=1-p_{22}(s,t)\).
If method.est=3 then \(p_{11}(s,t)\), \(p_{13}(s,t)\) and \(p_{23}(s,t)\) are estimated according to the following expressions:
\(p_{11}(s,t)=\frac{1-P(Z \leq t)}{1-P(Z \leq s)}\),
\(p_{13}(s,t)=\frac{P(Z>s,T \leq t)}{1-P(Z \leq s)}\),
\(p_{23}(s,t) =\frac{P(Z \leq s,s<T \leq t)}{P(Z \leq s)-P(T \leq s)}\).
Then, \(p_{12}(s,t)=1-p_{11}(s,t)-p_{13}(s,t)\) and \(p_{22}(s,t)=1-p_{23}(s,t)\).
If method.est=4 then \(p_{11}(s,t)\), \(p_{13}(s,t)\) and \(p_{23}(s,t)\) are estimated according to the following expressions:
\(p_{11}(s,t)=\frac{P(Z>t)}{P(Z>s)}\),
\(p_{13}(s,t)=\frac{P(Z>s,T \leq t)}{P(Z>s)}\),
\(p_{23}(s,t) =\frac{P(Z \leq s,s<T \leq t)}{P(Z \leq s,T>s)}\).
Then, \(p_{12}(s,t)=1-p_{11}(s,t)-p_{13}(s,t)\) and \(p_{22}(s,t)=1-p_{23}(s,t)\).
Amorim A. P., de Uña-Álvarez J., Meira Machado L. F. (2011). Presmoothing the transition probabilities in the illness-death model. Statistics and Probability Letters, 81(7), 797-806. tools:::Rd_expr_doi("10.1016/j.spl.2011.02.017")
Araújo A, Meira-Machado L, Roca-Pardiñas J (2014). TPmsm: Estimation of the Transition Probabilities in 3-State Models. Journal of Statistical Software, 62(4), 1-29. tools:::Rd_expr_doi("10.18637/jss.v062.i04")
Davison, A. C., Hinkley, D. V. (1997). Bootstrap Methods and their Application, Chapter 5, Cambridge University Press.
transAJ,
transIPCW,
transKMW,
transLIN,
transLS,
transPAJ.
# Set the number of threads
nth <- setThreadsTP(2);
# Create survTP object
data(heartTP);
heartTP_obj <- with( heartTP, survTP(time1, event1, Stime, event) );
# Compute transition probabilities
transKMPW(object=heartTP_obj, s=33, t=412);
# Compute transition probabilities with confidence band
transKMPW(object=heartTP_obj, s=33, t=412, conf=TRUE, conf.level=0.9,
method.boot="percentile", method.est=4);
# Restore the number of threads
setThreadsTP(nth);
Run the code above in your browser using DataLab