Obtain a confidence interval from coefficient draws

Source: R/utils.R

Returns a data frame with the columns 'mean', 'lower', and 'upper' reflecting the mean and lower and upper bounds of a confidence interval (quantiles) for every column in a data frame of draws

ci(df, level = 0.95)

Arguments

df

A data frame of draws with all numeric columns.

level

The sensitivity of the computed confidence interval (CI). Defaults to level = 0.95, reflecting a 95% CI.

Examples

library(logitr)

# Estimate a preference space model
mnl_pref <- logitr(
  data    = yogurt,
  outcome = "choice",
  obsID   = "obsID",
  pars    = c("price", "feat", "brand")
)
#> Running model...
#> Done!

# Obtain 10,000 draws of parameters from model
coefs <- coef(mnl_pref)
covariance <- vcov(mnl_pref)
coef_draws <- as.data.frame(MASS::mvrnorm(10^4, coefs, covariance))

# Compute a confidence interval
ci(coef_draws, level = 0.95)
#>                    mean      lower      upper
#> price        -0.3664219 -0.4146429 -0.3181950
#> feat          0.4904642  0.2513380  0.7239711
#> brandhiland  -3.7148090 -3.9958394 -3.4295248
#> brandweight  -0.6415058 -0.7477549 -0.5338245
#> brandyoplait  0.7341273  0.5760748  0.8942999