Outlier detection
Hypothesis tests · Method and input reference
Tests are two-sided. Normality reports Shapiro–Wilk and D’Agostino–Pearson where sample size permits. Outliers use Tukey 1.5×IQR fences; observations are not removed. Synthetic example: 120 baseline systolic blood pressures (mmHg), with between-person variation. Illustrative data, not clinical evidence.
Open calculator · Compare methods in the decision matrix
Required inputs
- Observations — Numeric observations; preserve the row and column structure described above.
Review missing values and data coding before running. Sample inputs are synthetic teaching examples.
Implementation
Lifescio formulas and summaries over NumPy arrays. Sample variances use ddof=1; custom formulas are described in the method help.
- numpy 2.2.6: numpy.quantile
These are pinned library versions; saved results record the versions actually used at runtime.