Confidence intervals for an ordered / inflated ordered fit

Wald intervals from the fitted covariance matrix (the interval for rho is formed on the \(\tanh^{-1}\) scale and transformed back, so it respects the (-1, 1) bounds), or – for a fit with se = "bootstrap" – percentile intervals from the bootstrap replicates.

Usage

# S3 method for class 'iord'
confint(object, parm, level = 0.95, type = c("wald", "percentile"), ...)

Arguments

object

An "iord" object fit with standard errors.

parm

Optional subset of coefficient names.

level

Confidence level (default 0.95).

type

"wald" (default) or "percentile" (requires se = "bootstrap"; quantiles of the replicate estimates on the natural scale).

...

Unused.

Value

A matrix of lower/upper bounds.

See also

vcov.iord(), summary.iord(), tidy.iord() (conf.int = TRUE).

Other inference methods: summary.iord(), vcov.iord()

Examples

data(pta)
m <- iop(flexibility ~ depth + democracy + gdp | gdp + democracy, data = pta, inflate = "top")
#> The inflation equation contains no covariate that is excluded from the outcome equation; the split is then identified by functional form alone. An exclusion restriction is advisable.
#> Warning: iop: the information matrix is nearly singular (reciprocal condition number 1.4e-19): some parameters are not separately identified (near-collinear covariates, a near-flat split direction, or a cutpoint at the boundary); their standard errors are unreliable.
confint(m)
#>                          2.5%       97.5%
#> depth              0.04220185  0.06695456
#> democracy         -0.52145169  0.12933945
#> gdp               -0.03889345  0.08824145
#> 0|1               -1.63981240  0.98935551
#> 1|2               -1.06279791  1.56376264
#> 2|3               -0.48678027  2.15209639
#> 3|4                0.15790426  2.82960388
#> infl_(Intercept)   3.79707575  5.07295576
#> infl_gdp           0.12962004  0.43075830
#> infl_democracy   -12.90280463 -7.79887626
confint(m, parm = c("depth", "infl_democracy"), level = 0.9)
#>                          5%         95%
#> depth            0.04419164  0.06496477
#> infl_democracy -12.49251630 -8.20916459
# \donttest{
mc <- iop(flexibility ~ depth + democracy + gdp | gdp + democracy, data = pta,
          inflate = "top", correlated = TRUE)
#> The inflation equation contains no covariate that is excluded from the outcome equation; the split is then identified by functional form alone. An exclusion restriction is advisable.
#> Warning: iop: the information matrix is nearly singular (reciprocal condition number 2.3e-19): some parameters are not separately identified (near-collinear covariates, a near-flat split direction, or a cutpoint at the boundary); their standard errors are unreliable.
confint(mc, parm = "rho")                      # formed on the atanh scale
#>          2.5%    97.5%
#> rho 0.1030959 0.846301
# }