Aggregate-level data mean and variance statistics

Computes the mean and variance of marginal treatment effects for aggregate-level trial data.

Usage

calc_ALD_stats(strategy, analysis_params)

Arguments

strategy

A list containing the strategy details, including the family distribution.

analysis_params

A list containing:

• ald Aggregate-level trial data

• ref_trt Treatment labels reference (common; e.g. placebo)

• ald_comp Treatment labels comparator

• scale A scaling parameter for the calculation. From "log_odds", "risk_difference", "log_relative_risk".

Value

A list containing:

mean

The marginal treatment effect mean.

var

The marginal treatment effect variance.

See also

marginal_treatment_effect(), marginal_variance()

Examples

strategy <- list(family = list(family = "binomial"))  # basic version

ald <- data.frame(trt = c("B","C","B","C"),
                  variable = c(NA, NA, "y", "y"),
                  statistic = c("N", "N", "sum", "sum"),
                  value = c(100, 100, 50, 60))

calc_ALD_stats(strategy = strategy,
               list(ald = ald,
                    ref_trt = "C",
                    ald_comp = "B",
                    scale = "log_odds"))
#> $mean
#> [1] -0.4054651
#> 
#> $var
#> [1] 0.08166667
#>