Ratify Attribute Assignment Results

Evaluates the quality of attribute assignment performed by a ReplicaAdder and stores validation diagnostics within the module.

Usage

# S4 method for class 'ReplicaAdder'
ratify(x, ...)

Arguments

x

A ReplicaAdder object.

...

Additional arguments passed to the generic method.

Value

An updated ReplicaAdder.

An updated ReplicaAdder.

Details

The ratify() method compares the observed distributions in the synthetic population with the expected distributions supplied via the reference contingency table.

Validation results are stored in the validation_results slot and include:

• goodness-of-fit statistics;

• p-values;

• warning flags; and

• detailed comparisons of observed and expected distributions.

These diagnostics can subsequently be explored using the validation and visualisation tools bundled with replica.

ReplicaAdder Method

Evaluates the quality of attribute assignment and stores validation diagnostics within the module.

The observed distributions in the synthetic population are compared with the expected distributions supplied by the contingency table.

Validation results are stored in:

x@validation_results

and include:

• z-squared statistics;

• p-values;

• warning flags; and

• detailed comparisons of observed and expected distributions.

ReplicaAdder

For a ReplicaAdder, ratify():

1. checks the assigned synthetic population;

2. compares observed and expected distributions;

3. calculates validation statistics;

4. stores validation results; and

5. returns the updated module.

Validation results are stored in:

x@validation_results

and may be inspected using:

names(x@validation_results)

See also

validate_synthetic_population_fit, plot_validation_distributions, plot_validation_differences, plot_validation_heatmap

ReplicaAdder, validate_synthetic_population_fit, plot_validation_distributions, plot_validation_differences, plot_validation_heatmap

Examples

if (FALSE) { # \dontrun{

adder <- enhance(
  adder
)

adder <- ratify(
  adder
)

adder@validation_results

} # }

if (FALSE) { # \dontrun{

adder <- ReplicaAdder(
  synth_pop = population,
  contingency = contingency,
  target_attribute = "education",
  group_by = c(
    "age_group",
    "gender"
  )
)

adder <- enhance(adder)

adder <- ratify(adder)

adder@validation_results
} # }