skpr (version 1.8.2)

get_optimality: Get optimality values

Description

Returns a list of optimality values (or one value in particular).

Note: The choice of contrast will effect the `G` efficiency value, and `gen_design()` and `eval_design()` by default set different contrasts (`contr.simplex` vs `contr.sum`).

Usage

get_optimality(output, optimality = NULL, calc_g = FALSE)

Value

A dataframe of optimality conditions. `D`, `A`, and `G` are efficiencies (value is out of 100). `T` is the trace of the information matrix, `E` is the minimum eigenvalue of the information matrix, `I` is the average prediction variance, and `Alias` is the trace of the alias matrix.

Arguments

output

The output of either gen_design or eval_design/eval_design_mc.

optimality

Default `NULL`. Return just the specific optimality requested.

calc_g

Default `FALSE`. Whether to calculate the g-efficiency.

Examples

# We can extract the optimality of a design from either the output of `gen_design()`
# or the output of `eval_design()`

factorialcoffee = expand.grid(cost = c(1, 2),
                             type = as.factor(c("Kona", "Colombian", "Ethiopian", "Sumatra")),
                             size = as.factor(c("Short", "Grande", "Venti")))

designcoffee = gen_design(factorialcoffee, ~cost + size + type, trials = 29,
                         optimality = "D", repeats = 100)

#Extract a list of all attributes
get_optimality(designcoffee)

#Get just one attribute
get_optimality(designcoffee,"D")

# Extract from `eval_design()` output
power_output = eval_design(designcoffee, model = ~cost + size + type,
                          alpha = 0.05, detailedoutput = TRUE)

get_optimality(power_output)