Specify a Gaussian-mixture truth

Usage

synthetic.truth.gaussian.mixture(
  centers,
  scales,
  weights,
  normalize = c("none", "sample.max"),
  evaluation.coordinates = "latent"
)

Arguments

centers: Component centers, one per row.
scales: Positive isotropic standard deviations or covariance matrices.
weights: Nonnegative mixture weights.
normalize: Normalization scope.
evaluation.coordinates: Coordinates on which to evaluate.

Value

A synthetic truth component.

Examples

synthetic.truth.gaussian.mixture(
  centers = rbind(-1, 1), scales = c(0.5, 0.5), weights = c(1, 1)
)
#> $kind
#> [1] "truth"
#> 
#> $family
#> [1] "gaussian.mixture"
#> 
#> $version
#> [1] 1
#> 
#> $parameters
#> $parameters$scope
#> [1] "population"
#> 
#> $parameters$estimand
#> [1] "Unnormalized Gaussian-mixture signal"
#> 
#> $parameters$evaluation.coordinates
#> [1] "latent"
#> 
#> $parameters$parameters
#> $parameters$parameters$centers
#>      [,1]
#> [1,]   -1
#> [2,]    1
#> 
#> $parameters$parameters$scales
#> [1] 0.5 0.5
#> 
#> $parameters$parameters$weights
#> [1] 0.5 0.5
#> 
#> $parameters$parameters$normalize
#> [1] "none"
#> 
#> 
#> 
#> attr(,"class")
#> [1] "synthetic_gaussian_mixture_truth" "synthetic_truth_component"       
#> [3] "synthetic_truth"                  "synthetic_component"