ATIP_MAHAL: Stata module for genuine multivariate outlier detection via Mahalanobis distance

Andres Talavera Cuya ()
Additional contact information
Andres Talavera Cuya: Direccion Nacional de Censos y Encuestas -- INEI Peru

Statistical Software Components from Boston College Department of Economics

Abstract: atip_mahal computes the multivariate Mahalanobis distance (2+ variables) within groups, detecting anomalous combinations that no univariate rule can see. By default it uses the Minimum Covariance Determinant estimator (Rousseeuw and Van Driessen 1999, Technometrics 41:212-223) instead of the classical covariance, which is highly robust to masking -- validated by simulation (1,000 datasets): at 20% contamination, recall stays at 100% with MCD versus 0% with the classical estimator. Includes an optional tolerance-ellipse graph (2 variables only) that can overlay both the classical and MCD ellipses to visualize the masking effect directly. A classic option reverts to the classical covariance estimator.

Language: Stata
Requires: Stata version 14
Keywords: outliers; Mahalanobis distance (search for similar items in EconPapers)
Date: 2026-08-02
Note: This module should be installed from within Stata by typing "ssc install atip_mahal". The module is made available under terms of the GPL v3 license.
References: Add references at CitEc
Citations:

Downloads: (external link)
http://fmwww.bc.edu/repec/bocode/a/atip_mahal.ado program code (text/plain)
http://fmwww.bc.edu/repec/bocode/a/atip_mahal.sthlp help file (text/plain)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:boc:bocode:s459848

Ordering information: This software item can be ordered from
http://repec.org/docs/ssc.php

Access Statistics for this software item

More software in Statistical Software Components from Boston College Department of Economics Boston College, 140 Commonwealth Avenue, Chestnut Hill MA 02467 USA. Contact information at EDIRC.
Bibliographic data for series maintained by Christopher F Baum ().