Hierarchical regularizers for mixed-frequency vector autoregressions

Alain Hecq; Marie Ternes; Ines Wilms

Hierarchical regularizers for mixed-frequency vector autoregressions

Research output: Working paper / Preprint › Working paper

Abstract

Mixed-frequency Vector AutoRegressions (MF-VAR) model the dynamics between variables recorded at different frequencies. However, as the number of series and high-frequency observations per low-frequency period grow, MF-VARs suffer from the "curse of dimensionality". We curb this curse through a regularizer that permits various hierarchical sparsity patterns by prioritizing the inclusion of coefficients according to the recency of the information they contain. Additionally, we investigate the presence of nowcasting relations by sparsely estimating the MF-VAR error covariance matrix. We study predictive Granger causality relations in a MF-VAR for the U.S. economy and construct a coincident indicator of GDP growth.

Original language	English
Publisher	Cornell University - arXiv
Number of pages	32
Publication status	Published - 2021

Publication series

Series	arXiv.org
Number	arXiv:2102.11780

JEL classifications

c32 - "Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"

Keywords

mixed-frequencies
vector autoregressions
high-dimensionality
group lasso
variable selection
coincident indicators

Access to Document

https://arxiv.org/pdf/2102.11780.pdf

Cite this

@techreport{04056d40815549eb9ac83de234aefda2,

title = "Hierarchical regularizers for mixed-frequency vector autoregressions",

abstract = "Mixed-frequency Vector AutoRegressions (MF-VAR) model the dynamics between variables recorded at different frequencies. However, as the number of series and high-frequency observations per low-frequency period grow, MF-VARs suffer from the {"}curse of dimensionality{"}. We curb this curse through a regularizer that permits various hierarchical sparsity patterns by prioritizing the inclusion of coefficients according to the recency of the information they contain. Additionally, we investigate the presence of nowcasting relations by sparsely estimating the MF-VAR error covariance matrix. We study predictive Granger causality relations in a MF-VAR for the U.S. economy and construct a coincident indicator of GDP growth.",

keywords = "mixed-frequencies, vector autoregressions, high-dimensionality, group lasso, variable selection, coincident indicators",

author = "Alain Hecq and Marie Ternes and Ines Wilms",

note = "FRED-MD/QD publicly available at https://research.stlouisfed.org/econ/mccracken/fred-databases/",

year = "2021",

language = "English",

series = "arXiv.org",

number = "arXiv:2102.11780",

publisher = "Cornell University - arXiv",

address = "United States",

type = "WorkingPaper",

institution = "Cornell University - arXiv",

}

TY - UNPB

T1 - Hierarchical regularizers for mixed-frequency vector autoregressions

AU - Hecq, Alain

AU - Ternes, Marie

AU - Wilms, Ines

N1 - FRED-MD/QD publicly available at https://research.stlouisfed.org/econ/mccracken/fred-databases/

PY - 2021

Y1 - 2021

N2 - Mixed-frequency Vector AutoRegressions (MF-VAR) model the dynamics between variables recorded at different frequencies. However, as the number of series and high-frequency observations per low-frequency period grow, MF-VARs suffer from the "curse of dimensionality". We curb this curse through a regularizer that permits various hierarchical sparsity patterns by prioritizing the inclusion of coefficients according to the recency of the information they contain. Additionally, we investigate the presence of nowcasting relations by sparsely estimating the MF-VAR error covariance matrix. We study predictive Granger causality relations in a MF-VAR for the U.S. economy and construct a coincident indicator of GDP growth.

AB - Mixed-frequency Vector AutoRegressions (MF-VAR) model the dynamics between variables recorded at different frequencies. However, as the number of series and high-frequency observations per low-frequency period grow, MF-VARs suffer from the "curse of dimensionality". We curb this curse through a regularizer that permits various hierarchical sparsity patterns by prioritizing the inclusion of coefficients according to the recency of the information they contain. Additionally, we investigate the presence of nowcasting relations by sparsely estimating the MF-VAR error covariance matrix. We study predictive Granger causality relations in a MF-VAR for the U.S. economy and construct a coincident indicator of GDP growth.

KW - mixed-frequencies

KW - vector autoregressions

KW - high-dimensionality

KW - group lasso

KW - variable selection

KW - coincident indicators

M3 - Working paper

T3 - arXiv.org

BT - Hierarchical regularizers for mixed-frequency vector autoregressions

PB - Cornell University - arXiv

ER -