Variational Bayesian functional PCA

A Bayesian approach to analyze the modes of variation in a set of curves is suggested. It is based on a generative model thus allowing for noisy and sparse observations of curves. A Demmler-Reinsch(-type) basis is used to enforce smoothness of the latent ('eigen')functions. Inference, including estimation, error assessment and model choice, particularly the choice of the number of eigenfunctions and their degree of smoothness, is derived from a variational approximation of the posterior distribution. The proposed analysis is illustrated with simulated and real data.

[1]  P. Hall,et al.  Properties of principal component methods for functional and longitudinal data analysis , 2006, math/0608022.

[2]  Ana M. Aguilera,et al.  Forecasting time series by functional PCA. Discussion of several weighted approaches , 1999, Comput. Stat..

[3]  Ana M. Aguilera,et al.  Forecasting PC-ARIMA Models for Functional Data , 2002, COMPSTAT.

[4]  Ana M. Aguilera,et al.  AN APPROXIMATED PRINCIPAL COMPONENT PREDICTION MODEL FOR CONTINUOUS-TIME STOCHASTIC PROCESSES , 1997 .

[5]  Models for Multivariate Data Analysis , 1994 .

[6]  Neil H. Timm,et al.  Multivariate Reduced-Rank Regression , 1999, Technometrics.

[7]  G. Wahba.  Spline models for observational data , 1990 .

[8]  Hans-Georg Müller,et al.  Classification using functional data analysis for temporal gene expression data , 2006, Bioinform..

[9]  H. Muller,et al.  Generalized functional linear models , 2005, math/0505638.

[10]  A. M. Aguilera,et al.  Principal component estimation of functional logistic regression: discussion of two different approaches , 2004 .

[11]  A. M. Aguilera,et al.  Modeling environmental data by functional principal component logistic regression , 2005 .

[12]  F. Yao,et al.  Penalized spline models for functional principal component analysis , 2006 .

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

[14]  P. Sarda,et al.  SPLINE ESTIMATORS FOR THE FUNCTIONAL LINEAR MODEL , 2003 .

[15]  Neil D. Lawrence,et al.  Variational inference for Student-t models: Robust Bayesian interpolation and generalised component analysis , 2005, Neurocomputing.

[16]  Ana M. Aguilera,et al.  Forecasting with unequally spaced data by a functional principal component approach , 1999 .

[17]  B. Bakshi,et al.  Bayesian principal component analysis , 2002 .

[18]  R. A. Choudrey.  FLEXIBLE BAYESIAN INDEPENDENT COMPONENT ANALYSIS FOR BLIND SOURCE SEPARATION , 2001 .

[19]  Hagai Attias,et al.  Independent Factor Analysis , 1999, Neural Computation.

[20]  Zhihua Zhang,et al.  Bayesian Inference on Principal Component Analysis Using Reversible Jump Markov Chain Monte Carlo , 2004, AAAI.

[21]  John A. Rice,et al.  Displaying the important features of large collections of similar curves , 1992 .

[22]  Xin Zhao,et al.  The functional data analysis view of longitudinal data , 2004 .

[23]  Václav Smídl,et al.  On Bayesian principal component analysis , 2007, Comput. Stat. Data Anal..

[24]  A. Linde.  PCA-based dimension reduction for splines , 2003 .

[25]  Michael A. West,et al.  BAYESIAN MODEL ASSESSMENT IN FACTOR ANALYSIS , 2004 .

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

[27]  Linear methods for regression and classification with functional data , 2007 .

[28]  Simon J. Godsill,et al.  Blind Separation of Sparse Sources Using Jeffrey's Inverse Prior and the EM Algorithm , 2006, ICA.

[29]  Bernard W. Silverman,et al.  Incorporating parametric effects into functional principal components analysis , 1995 .

[30]  Christopher M. Bishop,et al.  Bayesian PCA , 1998, NIPS.

[31]  Claude Manté,et al.  Principal component analysis of measures, with special emphasis on grain-size curves , 2007, Comput. Stat. Data Anal..

[32]  P. R. Bouzas,et al.  Functional approach to the random mean of a compound Cox process , 2007, Comput. Stat..

[33]  Philippe Besse,et al.  Simultaneous non-parametric regressions of unbalanced longitudinal data , 1997 .

[34]  A. Linde,et al.  Splines from a Bayesian point of view , 1995 .

[35]  Thomas A. Louis,et al.  Modeling the Labeling Index Distribution: An Application of Functional Data Analysis , 1995 .

[36]  Mariano J. Valderrama,et al.  An overview to modelling functional data , 2007, Comput. Stat..

[37]  Tom Minka,et al.  Automatic Choice of Dimensionality for PCA , 2000, NIPS.

[38]  Rob J. Hyndman,et al.  Robust forecasting of mortality and fertility rates: A functional data approach , 2007, Comput. Stat. Data Anal..

[39]  Ana M. Aguilera,et al.  Stochastic modelling for evolution of stock prices by means of functional principal component analysis , 1999 .

[40]  Ana M. Aguilera,et al.  Modelling the mean of a doubly stochastic Poisson process by functional data analysis , 2006, Comput. Stat. Data Anal..

[41]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

[42]  H. Cardot.  Nonparametric estimation of smoothed principal components analysis of sampled noisy functions , 2000 .

[43]  Ana M. Aguilera,et al.  Forecasting binary longitudinal data by a functional PC-ARIMA model , 2008, Comput. Stat. Data Anal..

[44]  Jeng-Min Chiou,et al.  Diagnostics for functional regression via residual processes , 2007, Comput. Stat. Data Anal..

[45]  Simon J. Godsill,et al.  Variational and stochastic inference for Bayesian source separation , 2007, Digit. Signal Process..

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

[47]  Ana M. Aguilera,et al.  Computational considerations in functional principal component analysis , 2007, Comput. Stat..

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

[49]  A. U.S.,et al.  Hierarchical Models for Assessing Variability among Functions , 2005 .

[50]  Michael E. Tipping,et al.  Probabilistic Principal Component Analysis , 1999 .

[51]  Bo Wang,et al.  Convergence and Asymptotic Normality of Variational Bayesian Approximations for Expon , 2004, UAI.

[52]  Angelika van der Linde.  Estimating the smoothing parameter in generalized spline-based regression , 2001, Comput. Stat..

[53]  Badih Ghattas,et al.  Classifying densities using functional regression trees: Applications in oceanology , 2007, Comput. Stat. Data Anal..

[54]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[55]  James O. Ramsay,et al.  Applied Functional Data Analysis: Methods and Case Studies , 2002 .

[56]  Ana M. Aguilera,et al.  Functional Principal Components Analysis by Choice of Norm , 1999 .

[57]  Angelika van der Linde.  Estimating the smoothing parameter in generalized spline-based regression , 2001, Comput. Stat..

[58]  K. J. Utikal,et al.  Inference for Density Families Using Functional Principal Component Analysis , 2001 .

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

[60]  Daniel Gervini,et al.  Free‐knot spline smoothing for functional data , 2006 .

[61]  R. M. Fernández-Alcalá,et al.  Functional estimation incorporating prior correlation information , 2007, Comput. Stat..

[62]  E. Ziegel,et al.  Proceedings in Computational Statistics , 1998 .

[63]  Wenceslao González-Manteiga,et al.  Statistics for Functional Data , 2007, Comput. Stat. Data Anal..

[64]  Ana M. Aguilera,et al.  Using principal components for estimating logistic regression with high-dimensional multicollinear data , 2006, Comput. Stat. Data Anal..

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

[66]  Charles M. Bishop.  Variational principal components , 1999 .

[67]  Ana M. Aguilera,et al.  Functional PLS logit regression model , 2007, Comput. Stat. Data Anal..

[68]  Hans-Georg Ller,et al.  Functional Modelling and Classification of Longitudinal Data. , 2005 .

[69]  John A. Rice,et al.  FUNCTIONAL AND LONGITUDINAL DATA ANALYSIS: PERSPECTIVES ON SMOOTHING , 2004 .