DeepAI AI Chat
Log In Sign Up

A robust approach for principal component analyisis

In this paper we analyze different ways of performing principal component analysis throughout three different approaches: robust covariance and correlation matrix estimation, projection pursuit approach and non-parametric maximum entropy algorithm. The objective of these approaches is the correction of the well known sensitivity to outliers of the classical method for principal component analysis. Due to their robustness, they perform very well in contaminated data, while the classical approach fails to preserve the characteristics of the core information.


page 1

page 2

page 3

page 4


A Bias Trick for Centered Robust Principal Component Analysis

Outlier based Robust Principal Component Analysis (RPCA) requires center...

Sparse Principal Component Analysis via Variable Projection

Sparse principal component analysis (SPCA) has emerged as a powerful tec...

On the Noise-Information Separation of a Private Principal Component Analysis Scheme

In a survey disclosure model, we consider an additive noise privacy mech...

Auto-associative models, nonlinear Principal component analysis, manifolds and projection pursuit

In this paper, auto-associative models are proposed as candidates to the...

Best Pair Formulation & Accelerated Scheme for Non-convex Principal Component Pursuit

The best pair problem aims to find a pair of points that minimize the di...

Robust Parameter Estimation for the Lee-Carter Model: A Probabilistic Principal Component Approach

As a traditional and widely-adopted mortality rate projection technique,...

Initialization of Self-Organizing Maps: Principal Components Versus Random Initialization. A Case Study

The performance of the Self-Organizing Map (SOM) algorithm is dependent ...