If I want to draw a calibration curve comparing predicted probabilities to actual events, some (e.g., Frank Harrell) suggest lowess regression of events against predicted probabilities as the preferred method, but also mention splines as another possibility. When fitting a calibration curve using splines, my intuition due to the is that I should fit a generalized additive model from a binomial family rather than assume a Gaussian error structure. When I do this, I often get wildly different results than using Gaussian GAM or lowess. Usually, the calibration curves using a binomial GAM looks much worse.

My question is what semi-parametric or non-parametric method to use when fitting calibration curves and why.

| cite | improve this question | | | | |

I asked this question on the Data Methods Discussion Forum and got an answer from Frank Harrell that restricted cubic splines work well, with reference to a paper on the topic.

| cite | improve this answer | | | | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.