Skip to main content
. 2023 Jul 19;65(8):2200302. doi: 10.1002/bimj.202200302

FIGURE 4.

FIGURE 4

Instability of estimated risks across 1000 example prediction models for each of nine individuals with true risks between 0.1 and 0.9, for model development sample sizes (nD ) of 50, 100, 385, 500, 1000, and 5000 participants: each example model was produced from a logistic regression (LR) with a LASSO penalty fitted to a different random sample of individuals from a population with a true overall risk of 0.5, considering 1 genuine predictor (X~N(0,4)) and 10 noise variables (Z1, …, Z10 ~ N(0,1)).