Z-standardise a dataframe of back-transformed Out-Of-Sample Predictions

Standardizes out-of-sample predictions by computing the Fisher's Z transformed Correlation Coefficient from analysts' out-of-sample prediction estimates and corresponding standard error.

Usage

pred_to_Z(back_transformed_data, params)

Arguments

back_transformed_data

a dataframe or tibble with the columns "estimate" and "se.fit", containing yi and SE\(yi\) values respectively

response_variable_name

a character vector

Value

A tibble of standardised-out-of-sample predictions on the Z-scale, with columns Z, VZ, lower and upper, and the original columns fro back_transformed_data that were not used / updated in the transformation.

Details

This function is used to standardize out-of-sample predictions on the response scale to the Z-scale. pred_to_Z() expects estimates to be on the response scale, not the link scale.

The function computes the Z-score and VZ-score for each out-of-sample prediction estimate and its corresponding standard error using Z_VZ_preds().

See also

Equivalent tolog_transform_yi() in terms of workflow data hierarchy.