Builds one or more texreg objects from a koma_estimate, so results
can be rendered with texreg::screenreg() or similar helpers.
Usage
extract.koma_estimate(
model,
variables = NULL,
central_tendency = "mean",
ci_low = 5,
ci_up = 95,
digits = 2,
...
)Arguments
- model
A
koma_estimateobject.- variables
Optional character vector of endogenous variables to include. Defaults to all variables in
model$estimates.- central_tendency
Central tendency used when summarizing estimates (e.g., "mean", "median"). Defaults to "mean".
- ci_low
Lower bound (percent) for credible intervals. Defaults to 5.
- ci_up
Upper bound (percent) for credible intervals. Defaults to 95.
- digits
Number of digits to round numeric values. Defaults to 2.
- ...
Unused. Included for
texreg::extract()compatibility.
See also
summary.koma_estimate for
summary output with optional texreg formatting.
Examples
if (requireNamespace("texreg", quietly = TRUE)) {
data("simulated_sem")
set.seed(11)
fit <- estimate(
ts_data = simulated_sem$ts_data,
sys_eq = simulated_sem$sys_eq,
dates = simulated_sem$dates,
options = list(gibbs = list(ndraws = 10))
)
texreg::extract(fit, variables = "consumption")
}
#>
#> ── Gibbs Sampler Settings ──────────────────────────────────────────────────────
#>
#> ── System Wide Settings ──
#> • Number of draws (`ndraws`): 10
#> • Burn-in ratio (`burnin_ratio`): 0.5
#> • Burn-in (`burnin`): 5
#> • Store frequency (`nstore`): 1
#> • Number of saved draws (`nsave`): 5
#> • Tau (`tau`): 1.1
#>
#>
#> ── Estimation ──────────────────────────────────────────────────────────────────
#> Model name: KOMANo decimal places were defined for the GOF statistics.
#>
#> coef. lower CI upper CI
#> constant 1.2315823 1.0903710 1.5150503
#> consumption.L(1) 0.4887485 0.4242359 0.5610555
#> consumption.L(2) 0.2172775 0.1705978 0.2699137
#> gdp -0.3345256 -0.3900799 -0.2511942
#>
#> No GOF block defined.