What you see may not be what you get¶
Babyak, M. A. (2004). What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models. Psychosomatic Medicine, 66(3), 411-421.
Cited by¶
1 citation across 1 artifact.
Each citation links to the sentence it supports in the citing article.
Mechanisms¶
- Decay-Curve Fit and Half-Life Estimate
- It misleads when the fitted regime does not hold: extrapolating an exponential past a plateau or a second, faster driver produces a confident wrong date, and a clean R² can mask an overfit
This sourceShows that overfit models can achieve persuasive in-sample R² values from noise yet fail to replicate in new samples.
- It misleads when the fitted regime does not hold: extrapolating an exponential past a plateau or a second, faster driver produces a confident wrong date, and a clean R² can mask an overfit
Verification¶
This reference passed the adversarial substantiation pipeline: it was checked to exist and to support the claim it is attached to. See how references were verified.
Registry ID ref:869e75f0ea47 · see in the full table