One-row summary of a fit
Value
A one-row tibble. nobs is the number of subjects, which is the
unit the asymptotics are in. No AIC or logLik column is reported: the
criterion these estimators optimise is a kernel-weighted pseudo-likelihood,
so an information criterion built from it would not mean what it usually
means.
Examples
set.seed(1)
d <- sim_async_data(n = 150)
fit <- kee_async(d$y, d$x, y ~ x, id = id, time = time, h = 0.3)
glance(fit)
#> # A tibble: 1 × 7
#> nobs nterms h npair link one_sided convergence
#> <int> <int> <dbl> <int> <chr> <lgl> <int>
#> 1 150 2 0.3 1863 identity FALSE 0