A Method of L1-Norm Principal Component Analysis for Functional Data

Recently, with the popularization of intelligent terminals, research on intelligent big data has been paid more attention. Among these data, a kind of intelligent big data with functional characteristics, which is called functional data, has attracted attention. Functional data principal component analysis (FPCA), as an unsupervised machine learning method, plays a vital role in the analysis of functional data. FPCA is the primary step for functional data exploration, and the reliability of FPCA plays an important role in subsequent analysis. However, classical L2-norm functional data principal component analysis (L2-norm FPCA) is sensitive to outliers. Inspired by the multivariate data L1-norm principal component analysis methods, we propose an L1-norm functional data principal component analysis method (L1-norm FPCA). Because the proposed method utilizes L1-norm, the L1-norm FPCs are less sensitive to the outliers than L2-norm FPCs which are the characteristic functions of symmetric covariance operator. A corresponding algorithm for solving the L1-norm maximized optimization model is extended to functional data based on the idea of the multivariate data L1-norm principal component analysis method. Numerical experiments show that L1-norm FPCA proposed in this paper has a better robustness than L2-norm FPCA, and the reconstruction ability of the L1-norm principal component analysis to the original uncontaminated functional data is as good as that of the L2-norm principal component analysis.

[1]  Panos P. Markopoulos,et al.  Optimal Algorithms for L1-subspace Signal Processing , 2014, IEEE Transactions on Signal Processing.

[2]  Ronaldo Dias,et al.  Functional data clustering via hypothesis testing k-means , 2019, Comput. Stat..

[3]  Jorge Mateu,et al.  Continuous Time-Varying Kriging for Spatial Prediction of Functional Data: An Environmental Application , 2010 .

[4]  Feiping Nie,et al.  Robust Principal Component Analysis with Non-Greedy l1-Norm Maximization , 2011, IJCAI.

[5]  R. Fraiman,et al.  Trimmed means for functional data , 2001 .

[6]  Vicente Zarzoso,et al.  On the Link Between L1-PCA and ICA , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Catherine A. Sugar,et al.  Principal component models for sparse functional data , 1999 .

[8]  Park Young Woong,et al.  Iteratively Reweighted Least Squares Algorithms for L1-Norm Principal Component Analysis , 2016 .

[9]  Piotr Kokoszka,et al.  Inference for Functional Data with Applications , 2012 .

[10]  M. Hallin,et al.  Dynamic functional principal components , 2015 .

[11]  H. Müller,et al.  Functional Data Analysis for Sparse Longitudinal Data , 2005 .

[12]  Panos P. Markopoulos,et al.  Adaptive L1-Norm Principal-Component Analysis With Online Outlier Rejection , 2018, IEEE Journal of Selected Topics in Signal Processing.

[13]  J. Dauxois,et al.  Asymptotic theory for the principal component analysis of a vector random function: Some applications to statistical inference , 1982 .

[14]  Frédéric Ferraty,et al.  Conditional Quantiles for Dependent Functional Data with Application to the Climatic El Niño Phenomenon , 2005 .

[15]  S. Ghigo,et al.  Analysis of air quality monitoring networks by functional clustering , 2008 .

[16]  B. Silverman,et al.  Estimating the mean and covariance structure nonparametrically when the data are curves , 1991 .

[17]  Liming Liu,et al.  A method for detecting outliers in functional data , 2017, IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society.

[18]  José A. Vilar,et al.  Functional ANOVA starting from discrete data: an application to air quality data , 2013, Environmental and Ecological Statistics.

[19]  B. Silverman,et al.  Functional Data Analysis , 1997 .

[20]  Daniel R. Kowal.  Integer‐valued functional data analysis for measles forecasting , 2019, Biometrics.

[21]  J. Ramsay,et al.  Some Tools for Functional Data Analysis , 1991 .

[22]  Alois Kneip,et al.  Common Functional Principal Components , 2006 .

[23]  B. Silverman,et al.  Smoothed functional principal components analysis by choice of norm , 1996 .

[24]  Tsagkarakis Nicholas,et al.  On the L1-Norm Approximation of a Matrix by Another of Lower Rank , 2016 .

[25]  J. Rubinstein,et al.  Hilbert-space Karhunen-Loève transform with application to image analysis. , 1999, Journal of the Optical Society of America. A, Optics, image science, and vision.

[26]  Ivan G. Guardiola,et al.  A Functional Data Analysis Approach to Traffic Volume Forecasting , 2018, IEEE Transactions on Intelligent Transportation Systems.

[27]  Sd Pezzulli,et al.  Some properties of smoothed principal components analysis for functional data , 1993 .

[28]  Henry W. Altland,et al.  Applied Functional Data Analysis , 2003, Technometrics.

[29]  Z. Q. John Lu,et al.  Nonparametric Functional Data Analysis: Theory And Practice , 2007, Technometrics.

[30]  Frédéric Ferraty,et al.  Nonparametric models for functional data, with application in regression, time series prediction and curve discrimination , 2004 .

[31]  Gillian Z Heller,et al.  Functional data analysis with application to periodically stimulated foetal heart rate data. II: Functional logistic regression , 2002, Statistics in medicine.

[32]  T. Auton.  Applied Functional Data Analysis: Methods and Case Studies , 2004 .

[33]  M. Febrero,et al.  Outlier detection in functional data by depth measures, with application to identify abnormal NOx levels , 2008 .

[34]  Panos P. Markopoulos,et al.  Efficient L1-Norm Principal-Component Analysis via Bit Flipping , 2016, IEEE Transactions on Signal Processing.

[35]  B. Mallick,et al.  Bayesian Hierarchical Spatially Correlated Functional Data Analysis with Application to Colon Carcinogenesis , 2008, Biometrics.

[36]  Nojun Kwak,et al.  Principal Component Analysis Based on L1-Norm Maximization , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[37]  P. Hall,et al.  On properties of functional principal components analysis , 2006 .

[38]  R. Fraiman,et al.  Kernel-based functional principal components ( , 2000 .

[39]  Thaddeus Tarpey,et al.  Clustering Functional Data , 2003, J. Classif..