Robust regularized singular value decomposition with application to mortality data

DOI10.1214/13-AOAS649zbMATH Open1454.62189arXiv1311.7480OpenAlexW3103431687MaRDI QIDQ386741FDOQ386741

Authors: Lingsong Zhang, Haipeng Shen, Jianhua Z. Huang 

Publication date: 10 December 2013

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: We develop a robust regularized singular value decomposition (RobRSVD) method for analyzing two-way functional data. The research is motivated by the application of modeling human mortality as a smooth two-way function of age group and year. The RobRSVD is formulated as a penalized loss minimization problem where a robust loss function is used to measure the reconstruction error of a low-rank matrix approximation of the data, and an appropriately defined two-way roughness penalty function is used to ensure smoothness along each of the two functional domains. By viewing the minimization problem as two conditional regularized robust regressions, we develop a fast iterative reweighted least squares algorithm to implement the method. Our implementation naturally incorporates missing values. Furthermore, our formulation allows rigorous derivation of leave-one-row/column-out cross-validation and generalized cross-validation criteria, which enable computationally efficient data-driven penalty parameter selection. The advantages of the new robust method over nonrobust ones are shown via extensive simulation studies and the mortality rate application.

Full work available at URL: https://arxiv.org/abs/1311.7480

Recommendations

• The analysis of two-way functional data using two-way regularized singular value decompositions
• Sparse-smooth regularized singular value decomposition
• Statistical inference based on robust low-rank data matrix approximation
• Robust singular value decomposition analysis of microarray data
• Functional Singular Component Analysis

zbMATH Keywords

functional data analysisprincipal component analysiscross-validationrobustnesssmoothing splineGCV

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Factor analysis and principal components; correspondence analysis (62H25) Applications of statistics to biology and medical sciences; meta analysis (62P10) Robustness and adaptive procedures (parametric inference) (62F35) Functional data analysis (62R10)

Cites Work

• Functional data analysis.
• Nonparametric functional data analysis. Theory and practice.
• The analysis of two-way functional data using two-way regularized singular value decompositions
• Functional Data Analysis for Sparse Longitudinal Data
• Title not available (Why is that?)
• Title not available (Why is that?)
• Robust forecasting of mortality and fertility rates: a functional data approach
• Robust principal component analysis for functional data. (With comments)
• Applied functional data analysis. Methods and case studies
• Robust factor analysis.
• Forecasting functional time series
• Some results on Tchebycheffian spline functions and stochastic processes
• Robust functional estimation using the median and spherical principal components
• Least Median of Squares Regression
• Robust regularized singular value decomposition with application to mortality data
• Title not available (Why is that?)
• Robust functional principal components: a projection-pursuit approach
• Robust Statistics
• Robust Statistics
• Smoothed functional principal components analysis by choice of norm
• Detecting and handling outlying trajectories in irregularly sampled functional datasets
• Lower Rank Approximation of Matrices by Least Squares with Any Choice of Weights
• Functional principal components analysis via penalized rank one approximation
• Sparse logistic principal components analysis for binary data
• The functional singular value decomposition for bivariate stochastic processes
• Title not available (Why is that?)
• A spatio-temporal solution for the EEG/MEG inverse problem using group penalization methods
• Robust Singular Value Decompositions: A New Approach to Projection Pursuit
• Robust singular value decomposition analysis of microarray data
• Use of multiple singular value decompositions to analyze complex intracellular calcium ion signals
• Robust estimation of the self-similarity parameter in network traffic using wavelet transform

Cited In (14)

• Intergenerational actuarial fairness when longevity increases: amending the retirement age
• Addressing the life expectancy gap in pension policy
• Robust deep neural network estimation for multi-dimensional functional data
• A comparative study of singular spectrum analysis, neural network, ARIMA and exponential smoothing for monthly rainfall forecasting
• Testing for a unit root in Lee-Carter mortality model
• Inference pitfalls in Lee-Carter model for forecasting mortality
• Robust low-rank data matrix approximations
• Robust singular value decomposition with application to video surveillance background modelling
• Cauchy robust principal component analysis with applications to high-deimensional data sets
• Intraday forecasts of a volatility index: functional time series methods with dynamic updating
• Robust regularized singular value decomposition with application to mortality data
• Multilevel Matrix-Variate Analysis and its Application to Accelerometry-Measured Physical Activity in Clinical Populations
• Low-Rank Covariance Function Estimation for Multidimensional Functional Data
• The analysis of two-way functional data using two-way regularized singular value decompositions

Uses Software

• robustbase
• forecast
• fda (R)

This page was built for publication: Robust regularized singular value decomposition with application to mortality data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q386741)