Extract Least Square Means from a GEE Model

Usage

lsmeans(object, conf_level = 0.95, weights = "proportional", ...)

# S3 method for tern_gee_logistic
lsmeans(object, conf_level = 0.95, weights = "proportional", ...)

Arguments

object: (tern_gee)
result of fit_gee().
conf_level: (proportion)
confidence level
weights: (string)
type of weights to be used for the least square means, see emmeans::emmeans() for details.
...: additional arguments for methods

Value

A data.frame with least-square means and contrasts. Additional classes allow to dispatch downstream methods correctly, too.

Examples

df <- fev_data
df$AVAL <- rbinom(n = nrow(df), size = 1, prob = 0.5)
fit <- fit_gee(vars = vars_gee(arm = "ARMCD"), data = df)

lsmeans(fit)
#>   ARMCD  prop_est prop_est_se prop_lower_cl prop_upper_cl   n    or_est
#> 1   PBO 0.5349901  0.02202202     0.4916734     0.5777854 420        NA
#> 2   TRT 0.5254463  0.02338787     0.4795214     0.5709448 380 0.9624085
#>   or_lower_cl or_upper_cl  log_or_est log_or_lower_cl log_or_upper_cl
#> 1          NA          NA          NA              NA              NA
#> 2   0.7474451    1.239195 -0.03831627      -0.2910944       0.2144619
#>   conf_level
#> 1       0.95
#> 2       0.95

lsmeans(fit, conf_level = 0.90, weights = "equal")
#>   ARMCD  prop_est prop_est_se prop_lower_cl prop_upper_cl   n    or_est
#> 1   PBO 0.5349901  0.02202202     0.4916734     0.5777854 420        NA
#> 2   TRT 0.5254463  0.02338787     0.4795214     0.5709448 380 0.9624085
#>   or_lower_cl or_upper_cl  log_or_est log_or_lower_cl log_or_upper_cl
#> 1          NA          NA          NA              NA              NA
#> 2    0.778447    1.189844 -0.03831627      -0.2504544       0.1738218
#>   conf_level
#> 1        0.9
#> 2        0.9