search

datarichard / shit-happens-index

Estimating coefficients for measuring severity of psychological distress resulting from recent life events
Creative Commons Attribution 4.0 International
1 stars 0 forks source link

test the predictive value of the life event scores in a cross-validation sample #5

Open datarichard opened 1 year ago

datarichard commented 1 year ago

Currently the predictive value is estimated and reported in the same sample. However testing in an independent sample is a prerequisite for any risk model, and we should report the standard prediction metrics for discrimination, calibration and classification (see Tripod elaboration paper from 2015).

datarichard commented 1 year ago

Box C in the Tripod paper says a common aim of prediction model development is quantifying the incremental or added predictive value of a specific (e.g., newly acquired) predictor. This seems synonymous with our aim here