The origin and destination state pairs a fitted model covers. A multi-state fit returns more than one row; a single-transition fit returns one.
See also
Other rate models:
tda_constrain(),
tda_control(),
tda_dple(),
tda_rates(),
tda_split(),
tda_survivor(),
vcov.tda_fit()
Examples
set.seed(9)
mn <- 60
m1 <- data.frame(ts = 0, tf = round(rexp(mn, 0.1), 1) + 0.5, org = 0,
des = sample(c(0, 1, 2), mn, TRUE, c(.25, .45, .30)),
x = round(rnorm(mn), 2))
mi <- which(m1$des == 1)
m2 <- data.frame(ts = m1$tf[mi],
tf = m1$tf[mi] + round(rexp(length(mi), 0.08), 1) + 0.5,
org = 1, des = sample(c(1, 2), length(mi), TRUE, c(.4, .6)),
x = m1$x[mi])
md <- rbind(m1, m2)
# three transitions: 0->1, 0->2 (censored at 2), 1->2
mf <- tda_rate(Surv(ts, tf, org, des) ~ x, md, model = "exponential")
tda_transitions(mf)
#> Org Des
#> 1 0 1
#> 2 0 2
#> 3 1 2