Correspondence analysis: handling cell-wise outliers via the reconstitution algorithm

TL;DR

This paper introduces a reconstitution algorithm for correspondence analysis that effectively handles cell-wise outliers by reducing their influence without removing entire rows or columns, preserving more data integrity.

Contribution

The paper presents a novel reconstitution algorithm that addresses cell-wise outliers in CA, improving robustness while maintaining data structure.

Findings

Reconstitution algorithm reduces outlier impact effectively.

Compared with supplementary points and MacroPCA, it performs better.

Maintains more information in the analysis.

Abstract

Correspondence analysis (CA) is a popular technique to visualize the relationship between two categorical variables. CA uses the data from a two-way contingency table and is affected by the presence of outliers. The supplementary points method is a popular method to handle outliers. Its disadvantage is that the information from entire rows or columns is removed. However, outliers can be caused by cells only. In this paper, a reconstitution algorithm is introduced to cope with such cells. This algorithm can reduce the contribution of cells in CA instead of deleting entire rows or columns. Thus the remaining information in the row and column involved can be used in the analysis. The reconstitution algorithm is compared with two alternative methods for handling outliers, the supplementary points method and MacroPCA. It is shown that the proposed strategy works well.