High-dimensional Linear Regression for Dependent Data with Applications to Nowcasting

DOI10.5705/SS.202018.0044MaRDI QIDQ4986331zbMATH OpenOpenAlexFDO

Authors Yuefeng Han, Ruey S. Tsay

Publication date 27 April 2021

Published in STATISTICA SINICA (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1706.07899, https://semanticscholar.org/paper/d14efa60312db642e9464c544ed0a398df18c15b

zbMATH Keywords

consistencyforecastinghigh-dimensional time seriesLassomixed-frequency datamodel selectionnowcasting

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Abstract: Recent research has focused on $el{l}_1$ penalized least squares (Lasso) estimators for high-dimensional linear regressions in which the number of covariates $p$ is considerably larger than the sample size $n$. However, few studies have examined the properties of the estimators when the errors and/or the covariates are serially dependent. In this study, we investigate the theoretical properties of the Lasso estimator for a linear regression with a random design and weak sparsity under serially dependent and/or nonsubGaussian errors and covariates. In contrast to the traditional case, in which the errors are independent and identically distributed and have finite exponential moments, we show that $p$ can be at most a power of $n$ if the errors have only finite polynomial moments. In addition, the rate of convergence becomes slower owing to the serial dependence in the errors and the covariates. We also consider the sign consistency of the model selection using the Lasso estimator when there are serial correlations in the errors or the covariates, or both. Adopting the framework of a functional dependence measure, we describe how the rates of convergence and the selection consistency of the estimators depend on the dependence measures and moment conditions of the errors and the covariates. Simulation results show that a Lasso regression can be significantly more powerful than a mixed-frequency data sampling regression (MIDAS) and a Dantzig selector in the presence of irrelevant variables. We apply the results obtained for the Lasso method to nowcasting with mixed-frequency data, in which serially correlated errors and a large number of covariates are common. The empirical results show that the Lasso procedure outperforms the MIDAS regression and the autoregressive model with exogenous variables in terms of both forecasting and nowcasting.

Recommendations

• Prediction in abundant high-dimensional linear regression
• Model selection for high-dimensional linear regression with dependent observations
• High-dimensional predictive regression in the presence of cointegration
• Partial least squares prediction in high-dimensional regression
• Stable prediction in high-dimensional linear models
• High-dimensional autocovariance matrices and optimal linear prediction
• Adaptive Handling of Dependence in High-Dimensional Regression Modeling
• High dimensional forecasting via interpretable vector autoregression
• Estimation of high-dimensional seemingly unrelated regression models
• High-dimensional linear regression via implicit regularization

Cited in

(14)

• High-dimensional inference for linear model with correlated errors
• High dimensional generalized linear models for temporal dependent data
• Finite sample theory for high-dimensional functional/scalar time series with applications
• Penalized averaging of parametric and non-parametric quantile forecasts
• Modified LASSO estimators for time series regression models with dependent disturbances
• Regularized estimation of high‐dimensional vector autoregressions with weakly dependent innovations
• Comments on ``Data science, big data and statistics
• Time-varying forecast combination for high-dimensional data
• Simultaneous Decorrelation of Matrix Time Series
• High-dimensional data segmentation in regression settings permitting temporal dependence and non-Gaussianity
• Change-point inference in high-dimensional regression models under temporal dependence
• _1-penalized multinomial regression: estimation, inference, and prediction, with an application to risk factor identification for different dementia subtypes
• Model selection for unit-root time series with many predictors
• Lasso-driven inference in time and space

This page was built for publication: High-dimensional Linear Regression for Dependent Data with Applications to Nowcasting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4986331)