Home▸Data Science▸Grubbs Test for Outliers: Critical Value from the t Distribution, Masking and Iterative Rejection

Grubbs Test for Outliers: Critical Value from the t Distribution, Masking and Iterative Rejection

Test whether the most extreme value in a roughly normal sample is an outlier. Compute the Grubbs statistic, compare it with the critical value from the t distribution, remove the point and repeat. Plant one or several outliers, change the sample size, significance level and shift, and watch masking hide a pair of outliers from a single-pass test. A Blender-built plotting easel with table-value, normal-limit, exact-bound and Monte Carlo size checks built in.

Data Science3DModerate60 FPS📱 Mobile-adapted🟠 Blender⇄ 2D version
Grubbs Test for Outliers: Critical Value from the t Distribution, Masking and Iterative Rejection ↗ Open standalone

Test whether the most extreme value in a roughly normal sample is an outlier. Compute the Grubbs statistic, compare it with the critical value from the t distribution, remove the point and repeat. Plant one or several outliers, change the sample size, significance level and shift, and watch masking hide a pair of outliers from a single-pass test. A Blender-built plotting easel with table-value, normal-limit, exact-bound and Monte Carlo size checks built in.

⚙ Under the hood

Test whether the most extreme value in a roughly normal sample is an outlier. Compute the Grubbs statistic, compare it with the critical value from the t distribution, remove the point and repeat. Plant one or several outliers, change the sample size, significance level and shift, and watch masking hide a pair of outliers from a single-pass test. A Blender-built plotting easel with table-value, normal-limit, exact-bound and Monte Carlo size checks built in.

Grubbs TestOutlier DetectionMasking EffectStudent t DistributionHypothesis Testing

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)