get_optimality: Get optimality values

In skpr: Design of Experiments Suite: Generate and Evaluate Optimal Designs

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)

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.

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.

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)

skpr documentation built on July 9, 2023, 7:23 p.m.