Detecting deviating data cells

TL;DR

This paper introduces a novel method for detecting individual deviating cells in multivariate data, accounting for variable correlations, without restrictions on clean data proportion, and providing estimates for outlying values.

Contribution

It presents the first approach to identify outlying data cells in multivariate datasets, considering correlations and handling high-dimensional data without requiring many clean rows.

Findings

Effectively uncovers more structure than columnwise or rowwise methods.

Can diagnose reasons for outlying cells, aiding process control.

Provides estimates of expected outlying cell values.

Abstract

A multivariate dataset consists of $n$ cases in $d$ dimensions, and is often stored in an $n$ by $d$ data matrix. It is well-known that real data may contain outliers. Depending on the situation, outliers may be (a) undesirable errors which can adversely affect the data analysis, or (b) valuable nuggets of unexpected information. In statistics and data analysis the word outlier usually refers to a row of the data matrix, and the methods to detect such outliers only work when at least half the rows are clean. But often many rows have a few contaminated cell values, which may not be visible by looking at each variable (column) separately. We propose the first method to detect deviating data cells in a multivariate sample which takes the correlations between the variables into account. It has no restriction on the number of clean rows, and can deal with high dimensions. Other advantages are that it provides estimates of the `expected' values of the outlying cells, while imputing missing values at the same time. We illustrate the method on several real data sets, where it uncovers more structure than found by purely columnwise methods or purely rowwise methods. The proposed method can help to diagnose why a certain row is outlying, e.g. in process control. It may also serve as an initial step for estimating multivariate location and scatter matrices.