Skip to contents

Construct a Canonical Bounded-Discrete Cluster-Count Target

DPprior_target_K.Rd

Constructs and validates a canonical elicitation target for the occupied cluster count \(K_J\) on the exact support \(1,\ldots,J\). Supply exactly one uncertainty source: var_K, confidence, cv_K, K_interval, or target_pmf.

Usage

DPprior_target_K(
  J,
  mu_K = NULL,
  var_K = NULL,
  confidence = NULL,
  cv_K = NULL,
  K_interval = NULL,
  target_pmf = NULL,
  tolerance = 1e-09,
  root_control = NULL
)

Arguments

J

Integer design size and upper support point for \(K_J\).

mu_K

Optional target mean for \(K_J\).

var_K

Optional target variance for \(K_J\).

confidence

Optional qualitative confidence label.

cv_K

Optional coefficient of variation, defined as \(SD(K_J)/E(K_J)\).

K_interval

Optional explicit interval-target specification.

target_pmf

Optional strict PMF on the support \(1,\ldots,J\); a structural zero for \(K=0\) may be supplied as the first of J+1 entries.

tolerance

Positive numerical postcondition tolerance.

root_control

Optional named controls for interval-target root solving.

Value

A validated dpprior.target/1 dpprior_K_target object. The exact support and original request are retained alongside normalized and used values, derivation evidence, implied moments, tolerances, verification, and provenance. Pass the object unchanged as target_K to DPprior_fit.

See also

Other elicitation: DPprior_a1(), DPprior_a2_kl(), DPprior_a2_newton(), DPprior_dual(), DPprior_dual_hard(), DPprior_dual_soft(), DPprior_fit()

Examples

target <- DPprior_target_K(J = 50, mu_K = 5, var_K = 8)
target$used[c("mu_K", "var_K")]
#> $mu_K
#> [1] 5
#> 
#> $var_K
#> [1] 8
#> 
fit <- DPprior_fit(
  J = 50, target_K = target, check_diagnostics = FALSE
)