Principal component analysis for second-order stationary vector time series
From MaRDI portal
Abstract: We extend the principal component analysis (PCA) to second-order stationary vector time series in the sense that we seek for a contemporaneous linear transformation for a -variate time series such that the transformed series is segmented into several lower-dimensional subseries, and those subseries are uncorrelated with each other both contemporaneously and serially. Therefore those lower-dimensional series can be analysed separately as far as the linear dynamic structure is concerned. Technically it boils down to an eigenanalysis for a positive definite matrix. When is large, an additional step is required to perform a permutation in terms of either maximum cross-correlations or FDR based on multiple tests. The asymptotic theory is established for both fixed and diverging when the sample size tends to infinity. Numerical experiments with both simulated and real data sets indicate that the proposed method is an effective initial step in analysing multiple time series data, which leads to substantial dimension reduction in modelling and forecasting high-dimensional linear dynamical structures. Unlike PCA for independent data, there is no guarantee that the required linear transformation exists. When it does not, the proposed method provides an approximate segmentation which leads to the advantages in, for example, forecasting for future values. The method can also be adapted to segment multiple volatility processes.
Recommendations
- Principal component analysis using frequency components of multivariate time series
- Generalized principal component analysis for moderately non-stationary vector time series
- Moving dynamic principal component analysis for non-stationary multivariate time series
- L’analyse en composantes principales de variables non stationnaires
- scientific article; zbMATH DE number 2232705
Cites work
- A canonical analysis of multiple time series
- A direct estimation of high dimensional stationary vector autoregressions
- A well-conditioned estimator for large-dimensional covariance matrices
- An improved Bonferroni procedure for multiple tests of significance
- Covariance regularization by thresholding
- Determining the Number of Factors in Approximate Factor Models
- Dynamic orthogonal components for multivariate time series
- Elements of multivariate time series analysis
- Estimation of latent factors for high-dimensional time series
- Factor modeling for high-dimensional time series: inference for the number of factors
- Forecasting Using Principal Components From a Large Number of Predictors
- High dimensional stochastic regression with latent factors, endogeneity and nonlinearity
- scientific article; zbMATH DE number 4135256 (Why is no real title available?)
- scientific article; zbMATH DE number 47363 (Why is no real title available?)
- scientific article; zbMATH DE number 3502569 (Why is no real title available?)
- scientific article; zbMATH DE number 954235 (Why is no real title available?)
- scientific article; zbMATH DE number 2199188 (Why is no real title available?)
- scientific article; zbMATH DE number 3395169 (Why is no real title available?)
- Identifying a Simplifying Structure in Time Series
- Instrumental Variable Algorithms for Multiple Input Systems Described by Multiple Transfer Functions
- Modelling multiple time series via common factors
- Modelling Multivariate Volatilities via Conditionally Uncorrelated Components
- Multivariate time series analysis. With R and financial applications
- Nonlinear spectral density estimation: thresholding the correlogram
- Nonlinear time series. Nonparametric and parametric methods
- Probability and moment inequalities under dependence
- Temporal BYY learning for state space approach, hidden Markov model, and blind source separation
- Testing for high-dimensional white noise using maximum cross-correlations
- The Generalized Dynamic Factor Model
- The Simes Method for Multiple Hypothesis Testing With Positively Dependent Test Statistics
- The use of factor analysis in the statistical analysis of multiple time series
Cited in
(39)- Optimal dimension reduction for high-dimensional and functional time series
- Confidence regions for entries of a large precision matrix
- Mortality forecasting using factor models: time-varying or time-invariant factor loadings?
- Blind source separation for compositional time series
- Canonical correlation for principal components of time series
- Moving dynamic principal component analysis for non-stationary multivariate time series
- Dynamic orthogonal components for multivariate time series
- Time dependent principal component analysis
- ON VECTOR AUTOCORRELATIONS AND GENERALIZED SECOND-ORDER FUNCTIONS FOR TIME SERIES
- Identification of Shared Components in Large Ensembles of Time Series Using Dimension Reduction
- Approximation par moyennage de l'analyse en composantes principales d'un processus stochastique
- Determining the signal dimension in second order source separation
- Constrained Factor Models for High-Dimensional Matrix-Variate Time Series
- Multi-population mortality forecasting using tensor decomposition
- Forecasting multiple time series with one-sided dynamic principal components
- Dynamic functional principal components
- Orthogonal connectivity factorization: interpretable decomposition of variability in correlation matrices
- scientific article; zbMATH DE number 2232705 (Why is no real title available?)
- Factor Models for High-Dimensional Tensor Time Series
- Testing the martingale difference hypothesis in high dimension
- Sparse principal component analysis for high‐dimensional stationary time series
- Factor modeling of multivariate time series: a frequency components approach
- Simultaneous Decorrelation of Matrix Time Series
- Forecasting Conditional Covariance Matrices in High-Dimensional Time Series: A General Dynamic Factor Approach
- A regularized low tubal-rank model for high-dimensional time series data
- On the Modeling and Prediction of High-Dimensional Functional Time Series
- Blind source separation over space: an eigenanalysis approach
- Multiperiod Dynamic Portfolio Choice: When High Dimensionality Meets Return Predictability
- On the empirical spectral distribution of large wavelet random matrices based on mixed-Gaussian fractional measurements in moderately high dimensions
- Identification and estimation for matrix time-series CP-factor models
- Sparse factor model for high dimensional time series
- Envelope Matrix Autoregressive Models
- Forecast linear augmented projection with targeted components
- Moment bounds for large autocovariance matrices under dependence
- A kernel multiple change-point algorithm via model selection
- HDTSA
- Principal component analysis using frequency components of multivariate time series
- Generalized principal component analysis for moderately non-stationary vector time series
- Time-dependent frequency domain principal components analysis of multichannel non-stationary signals
This page was built for publication: Principal component analysis for second-order stationary vector time series
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q82525)
