Health economics · Analysis and reporting

External Economic-Model Validation Plot

Synthetic external observed and predicted risks expose bias and misfit outside a model's calibration exercise.

Download SVG Open code and full worked example

When to use it

Use independent validation targets, prespecified tolerances and uncertainty; agreement with calibration data is not external validation.

How to read it

Observed risks are .05,.10,.20,.30,.40 and predictions .06,.11,.22,.34,.38. Signed mean error is .012 and mean absolute error .020; these do not establish acceptability without context and uncertainty.

Worked example

Synthetic external observed and predicted risks expose bias and misfit outside a model's calibration exercise.

Result: Use independent validation targets, prespecified tolerances and uncertainty; agreement with calibration data is not external validation.

Produced by

Health-Economic Model Verification, Calibration, and Validation Health Economic Modeling Methods Using RWE

Reference: Primary method sources are listed in the linked producing-method entries. The displayed data are original synthetic teaching examples, not published patient results.