Home▸Data Science▸Regression Dilution: Noisy Predictor and Attenuated Slope (2D)

Regression Dilution: Noisy Predictor and Attenuated Slope (2D)

A scatter of a noisy predictor against the outcome with the true, OLS and Deming lines, plus the slope-versus-error curve beta times lambda. Same shared model as the 3D lab, with numerical checks.

Data Science2DAdvanced60 FPS📱 Mobile-adapted🟠 Blender❄️ Ice & Cold⇄ 3D version
Regression Dilution: Noisy Predictor and Attenuated Slope (2D) ↗ Open standalone

A partial dependence plot answers one question: on average, how does the model's prediction change as one feature moves, once every other feature is averaged out? This simulator renders a real two-feature response surface as a 3D bar landscape, samples a set of data instances on it, and builds the PDP curve for whichever feature you choose to sweep by averaging that surface across all the sampled instances — exactly the formula used by real SHAP/PDP tooling. Individual conditional expectation (ICE) curves are drawn underneath for every instance so you can see the average being built from real per-instance behaviour, and an interaction-strength control lets you watch the moment PDP's core assumption — that the swept feature doesn't interact with the others — starts to break down, with the ICE curves fanning out around a PDP average that increasingly misrepresents any single instance.

⚙ Under the hood

A scatter of a noisy predictor against the outcome with the true, OLS and Deming lines, plus the slope-versus-error curve beta times lambda. Same shared model as the 3D lab, with numerical checks.

Regression DilutionMeasurement ErrorAttenuation BiasDeming RegressionReliability Ratio

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)