An Algebraic Estimator for Large Spectral Density Matrices
From MaRDI portal
Publication:6154009
Abstract: We propose a new estimator of high-dimensional spectral density matrices, called UNshrunk ALgebraic Spectral Estimator (UNALSE), under the assumption of an underlying low rank plus sparse structure, as typically assumed in dynamic factor models. The UNALSE is computed by minimizing a quadratic loss under a nuclear norm plus norm constraint to control the latent rank and the residual sparsity pattern. The loss function requires as input the classical smoothed periodogram estimator and two threshold parameters, the choice of which is thoroughly discussed. We prove consistency of UNALSE as both the dimension and the sample size diverge to infinity, as well as algebraic consistency, i.e., the recovery of latent rank and residual sparsity pattern with probability one. The finite sample properties of UNALSE are studied by means of an extended simulation exercise as well as an empirical analysis of US macroeconomic data.
Cites work
- A large covariance matrix estimator under intermediate spikiness regimes
- A Singular Value Thresholding Algorithm for Matrix Completion
- An iterative thresholding algorithm for linear inverse problems with a sparsity constraint
- Arbitrage, Factor Structure, and Mean-Variance Analysis on Large Asset Markets
- ASYMPTOTIC THEORY FOR SPECTRAL DENSITY ESTIMATES OF GENERAL MULTIVARIATE TIME SERIES
- Automatic Statistical Analysis of Bivariate Nonstationary Time Series
- Band Spectral Regression with Trending Data
- Convergence of covariance and spectral density estimates for high-dimensional locally stationary processes
- Covariance regularization by thresholding
- Data-driven shrinkage of the spectral density matrix of a high-dimensional time series
- Determining the Number of Factors in the General Dynamic Factor Model
- DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
- Dynamic factor models with infinite-dimensional factor space: asymptotic analysis
- Economic Predictions With Big Data: The Illusion of Sparsity
- Factor models in high-dimensional time series: A time-domain approach
- For most large underdetermined systems of linear equations the minimal 𝓁1‐norm solution is also the sparsest solution
- Granger causality and path diagrams for multivariate time series
- Graphical interaction models for multivariate time series.
- scientific article; zbMATH DE number 1911755 (Why is no real title available?)
- Large covariance estimation by thresholding principal orthogonal complements. With discussion and authors' reply
- Latent variable graphical model selection via convex optimization
- Linear Regression in the Frequency Domain
- Long-run covariability
- Rank-Sparsity Incoherence for Matrix Decomposition
- Shrinkage estimation in the frequency domain of multivariate time series
- SLEX Analysis of Multivariate Nonstationary Time Series
- Sparsistency and rates of convergence in large covariance matrix estimation
- Spectral analysis of high-dimensional time series
- Structural shrinkage of nonparametric spectral estimators for multivariate time series
- Testing for Common Trends
- Testing for short- and long-run causality: a frequency-domain approach
- Testing hypotheses about the number of factors in large factor models
- TESTS FOR SEASONAL COINTEGRATION USING PRINCIPAL COMPONENTS
- The Generalized Dynamic Factor Model
- The generalized shrinkage estimator for the analysis of functional connectivity of brain signals
- Time series. Data analysis and theory.
- Whittle Pseudo-Maximum Likelihood Estimation for Nonstationary Time Series
Cited in
(5)- The dynamic, the static, and the weak: factor models and the analysis of high-dimensional time series
- Forecasting dynamic factor scores by UNALSE spectral density matrix estimator
- Statistical Inference for High-Dimensional Spectral Density Matrix
- Frequency Domain Statistical Inference for High-Dimensional Time Series
- General spatio-temporal factor models for high-dimensional random fields on a lattice
This page was built for publication: An Algebraic Estimator for Large Spectral Density Matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6154009)
