Skip to contents

Applies the 12 regressions of wet_share_fit(), floors each prediction at 0 and renormalises so the 12 shares sum to 1. A watershed whose 12 floored predictions are all 0 gets NA shares.

Usage

wet_share_predict(fit, d)

Arguments

fit

A wet_share_fit() result.

d

data.frame with the predictors named in fit.

Value

A matrix with one row per row of d and columns share_01 to share_12.