test.nm {TP.idm}

Graphical tool to check the Markov assumption

Description

It constructs a PP-plot which compares the transition probabilities reported by the non-Markovian and Aalen-Johansen estimators. Under the Markov assumption the PP-plot should fit the straight line y=x. When the Markov assumption holds the Aalen-Johansen is preferred since it provides a smaller standard error. If the Markov assumption is violated, the Aalen-Johansen may be inconsistent and therefore the non-Markovian method is recommended.

Usage

test.nm(data, s, t = "last")

Arguments

data

A data.frame including at least four columns named time1, event1, Stime and event, which correspond to disease free survival time, disease free survival indicator, time to death or censoring, and death indicator, respectively.

s

The current time for the transition probabilities to be computed.

t

The future time for the transition probabilities to be computed. Default is “last” which means the largest time among the uncensored entry times for the intermediate state and the final absorbing state.

Details

It constructs a PP-plot which compares the transition probabilities reported by the non-Markovian and Aalen-Johansen estimators. Under the Markov assumption the PP-plot should fit the straight line y=x. When the Markov assumption holds the Aalen-Johansen is preferred since it provides a smaller standard error. If the Markov assumption is violated, the Aalen-Johansen may be inconsistent and therefore the non-Markovian method is recommended. The PP-plot excludes P_{11}(s,t) since both estimators agree in this case. Also, the user-supplied s must be strictly positive, because the Markov assumption is not relevant for the estimation of occupation probabilities (s=0).

Examples

data(colonTP)

test.nm(colonTP, s = 0) 
# nothig is displayed since the Markov condition is not relevant 
# for the case s=0 (occupation probabilities) 

test.nm(colonTP, s = 365, t = 1095)