Prepare X and y for CTA using SDA-selected attributes

Returns a named list list(X_cta, y_cta) where X_cta contains only the SDA-selected attribute columns and y_cta is the full outcome vector (all observations, not just unresolved).

Usage

sda_to_cta_data(fit, X, y)

Arguments

fit

An sda_fit object.

X

Data frame of predictors (all observations).

y

Integer class vector (all observations).

Value

Named list with elements X_cta (data frame, selected columns only) and y_cta (integer vector, full length).

Details

This matches the Path B workflow from MPE Chapter 12: SDA identifies the attribute subset; MDSA/CTA receives the full sample with a constrained candidate frame. SDA resolution does not restrict which observations CTA sees.