Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

How to construct a multiple regression model for data with missing elements and outlying objects.

Ivana Stanimirova1, Sven Serneels, Pierre J Van Espen

  • 1Department of Chemometrics, The University of Silesia, Katowice, Poland.

Analytica Chimica Acta
|March 28, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Lipidomic profile of meningiomas harboring different NF2 mutation status.

Metabolomics : Official journal of the Metabolomic Society·2026
Same author

Microextraction Technologies as Exposomic Sensors for On-Site Environmental Air Monitoring of Volatile Organic Compounds: A Review of Commercially Available Technologies.

Molecules (Basel, Switzerland)·2026
Same author

Insights into diet, psychological distress, and personality traits among patients with lower-extremity lymphedema and overweight/obesity in comparison to patients with lifestyle-induced overweight/obesity and patients with normal body weight.

Obesity research & clinical practice·2025
Same author

Spatial variability of pollution source contributions during two (2012-2013 and 2018-2019) sampling campaigns at ten sites in Los Angeles basin.

Environmental pollution (Barking, Essex : 1987)·2024
Same author

Association between skin lymphangiogenesis parameters and arterial hypertension status in patients: An observational study.

Advances in clinical and experimental medicine : official organ Wroclaw Medical University·2024
Same author

Quantification and Detection of Ground Garlic Adulteration Using Fourier-Transform Near-Infrared Reflectance Spectra.

Foods (Basel, Switzerland)·2023
Same journal

Smartphone-assisted fluorescence and colorimetric dual-mode sensor for visual quantitative detection of nitrite and nitrate in real samples.

Analytica chimica acta·2026
Same journal

Folding integrated all-paper photoelectrochemical immunoassay using annealed ZnO for point-of-care detection of ferritin.

Analytica chimica acta·2026
Same journal

Dual-mode electrochemical-SERS detection of chloramphenicol based on dual-signal enhancement.

Analytica chimica acta·2026
Same journal

Multi-screening of beta-lactam antibiotics in milk based on Fe<sub>3</sub>O<sub>4</sub>@phage/bacteria system and aggregation induced emission luminogen.

Analytica chimica acta·2026
Same journal

A porous phosphate-rich β-cyclodextrin polymer for efficient and broad-spectrum enrichment of antibiotics.

Analytica chimica acta·2026
Same journal

Corrigendum to "LUMIN: A novel algorithm for automated mixture quantification using 1D <sup>1</sup>H NMR spectra" [Analytica Chimica Acta 1411 (2026) 345639].

Analytica chimica acta·2026
See all related articles

Robust regression techniques using expectation maximization effectively model data with missing values and outliers. This approach offers a reliable method for building accurate regression models, outperforming standard methods in challenging datasets.

Area of Science:

  • Statistics
  • Data Science
  • Machine Learning

Background:

  • Traditional regression models struggle with datasets containing missing values and outliers.
  • Robust statistical methods are essential for reliable data analysis when data quality is compromised.

Purpose of the Study:

  • To demonstrate the effectiveness of robust multiple regression techniques within the expectation maximization framework.
  • To compare the performance of partial least squares and partial robust M-regression models using expectation maximization.

Main Methods:

  • Implementation of robust multiple regression techniques using the expectation maximization algorithm.
  • Comparative analysis involving partial least squares and partial robust M-regression.
  • Validation on simulated datasets with varying percentages of missing data and outliers, and on a real-world dataset.

Related Experiment Videos

Main Results:

  • The expectation maximization framework successfully accommodates missing data and outliers in regression modeling.
  • Partial robust M-regression within the expectation maximization framework showed strong performance.
  • The proposed methods yielded satisfactory regression models, evaluated by trimmed root mean squared errors.

Conclusions:

  • Robust multiple regression techniques integrated with expectation maximization provide a powerful solution for modeling incomplete and outlier-prone data.
  • The demonstrated methodology enhances the reliability and accuracy of regression models in practical applications.
  • This approach is valuable for data scientists and statisticians dealing with imperfect datasets.