Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data

P. W. Nyam; K. E. Lasisi; I. A. Baba

Home / Current Issue / Paper 1724750

1724750 Vol 10 · Issue 5 Download Paper

Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data

P. W. Nyam K. E. Lasisi I. A. Baba

Subject area: Science,Engineering and Technology · Area of research: Statistical Machine Learning

Abstract

Fisher discriminant analysis (FDA) is sensitive to outliers, and when the number of variables p exceeds the number of observations n, the sample covariance matrix is singular.We propose a hybrid robust covariance estimator that combines cellwise detection and imputation (Detect Deviating Cells) with rowwise MRCD estimation, mixes the two components with a weight, and applies ridge shrinkage so that the estimator is invertible for p > n. The estimator is used in a modified Fisher rule.

Keywords

Fisher discriminant analysis; cellwise outliers; DDC; MRCD; ridge regularization; high-dimensional classification

How to cite this paper

P. W. Nyam, K. E. Lasisi, I. A. Baba "Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data" Iconic Research And Engineering Journals Volume 10 Issue 5 2026 Page 1081-1087

P. W. Nyam, K. E. Lasisi, I. A. Baba "Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data" Iconic Research And Engineering Journals, vol. 10, no. 5, Nov. 2026

P. W. Nyam, K. E. Lasisi, I. A. Baba (2026). Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data. Iconic Research And Engineering Journals, 10(5).

P. W. Nyam, K. E. Lasisi, I. A. Baba "Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data" Iconic Research And Engineering Journals, vol. 10, no. 5, Nov. 2026.

@article{1724750,
      author = {P. W. Nyam, K. E. Lasisi, I. A. Baba},
      title = {Hybrid Cellwise-Robust Fisher Discriminant Analysis for High-Dimensional Data},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {10},
      number = {5},
      pages = {1081-1087},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1724750.pdf},
      abstract = {Fisher discriminant analysis (FDA) is sensitive to outliers, and when the number of variables p exceeds the number of observations n, the sample covariance matrix is singular.We propose a hybrid robust covariance estimator that combines cellwise detection and imputation (Detect Deviating Cells) with rowwise MRCD estimation, mixes the two components with a weight, and applies ridge shrinkage so that the estimator is invertible for p > n. The estimator is used in a modified Fisher rule.},
      keywords = {Fisher discriminant analysis; cellwise outliers; DDC; MRCD; ridge regularization; high-dimensional classification},
      month = {November},
  }