Assessing predictive accuracy in panel data models with long-range dependence

Publikation: Working paperForskning

Standard

Assessing predictive accuracy in panel data models with long-range dependence. / Borup, Daniel; Christensen, Bent Jesper; Ergemen, Yunus Emre.

Aarhus : Institut for Økonomi, Aarhus Universitet, 2019.

Publikation: Working paperForskning

Harvard

APA

Borup, D., Christensen, B. J., & Ergemen, Y. E. (2019). Assessing predictive accuracy in panel data models with long-range dependence. Aarhus: Institut for Økonomi, Aarhus Universitet. CREATES Research Papers, Nr. 2019-04

CBE

MLA

Borup, Daniel, Bent Jesper Christensen, og Yunus Emre Ergemen Assessing predictive accuracy in panel data models with long-range dependence. Aarhus: Institut for Økonomi, Aarhus Universitet. (CREATES Research Papers; Journal nr. 2019-04). 2019., 59 s.

Vancouver

Author

Borup, Daniel ; Christensen, Bent Jesper ; Ergemen, Yunus Emre. / Assessing predictive accuracy in panel data models with long-range dependence. Aarhus : Institut for Økonomi, Aarhus Universitet, 2019. (CREATES Research Papers; Nr. 2019-04).

Bibtex

@techreport{ac0d79833af24cf68fe5d43061197cf8,
title = "Assessing predictive accuracy in panel data models with long-range dependence",
abstract = "This paper proposes tests of the null hypothesis that model-based forecastsare uninformative in panels, allowing for individual and interactive fixed effectsthat control for cross-sectional dependence, endogenous predictors, andboth short-range and long-range dependence. We consider a Diebold-Marianostyle test based on comparison of the model-based forecast and a nested nopredictability benchmark, an encompassing style test of the same null, and atest of pooled uninformativeness in the entire panel. A simulation study showsthat the encompassing style test is reasonably sized in finite samples, whereasthe Diebold-Mariano style test is oversized. Both tests have non-trivial localpower. The methods are applied to the predictive relation between economicpolicy uncertainty and future stock market volatility in a multi-country analysis.",
keywords = "Panel data, predictability, long-range dependence, Diebold-Mariano test, encompassing test",
author = "Daniel Borup and Christensen, {Bent Jesper} and Ergemen, {Yunus Emre}",
year = "2019",
month = "3",
day = "28",
language = "English",
series = "CREATES Research Papers",
publisher = "Institut for {\O}konomi, Aarhus Universitet",
number = "2019-04",
type = "WorkingPaper",
institution = "Institut for {\O}konomi, Aarhus Universitet",

}

RIS

TY - UNPB

T1 - Assessing predictive accuracy in panel data models with long-range dependence

AU - Borup, Daniel

AU - Christensen, Bent Jesper

AU - Ergemen, Yunus Emre

PY - 2019/3/28

Y1 - 2019/3/28

N2 - This paper proposes tests of the null hypothesis that model-based forecastsare uninformative in panels, allowing for individual and interactive fixed effectsthat control for cross-sectional dependence, endogenous predictors, andboth short-range and long-range dependence. We consider a Diebold-Marianostyle test based on comparison of the model-based forecast and a nested nopredictability benchmark, an encompassing style test of the same null, and atest of pooled uninformativeness in the entire panel. A simulation study showsthat the encompassing style test is reasonably sized in finite samples, whereasthe Diebold-Mariano style test is oversized. Both tests have non-trivial localpower. The methods are applied to the predictive relation between economicpolicy uncertainty and future stock market volatility in a multi-country analysis.

AB - This paper proposes tests of the null hypothesis that model-based forecastsare uninformative in panels, allowing for individual and interactive fixed effectsthat control for cross-sectional dependence, endogenous predictors, andboth short-range and long-range dependence. We consider a Diebold-Marianostyle test based on comparison of the model-based forecast and a nested nopredictability benchmark, an encompassing style test of the same null, and atest of pooled uninformativeness in the entire panel. A simulation study showsthat the encompassing style test is reasonably sized in finite samples, whereasthe Diebold-Mariano style test is oversized. Both tests have non-trivial localpower. The methods are applied to the predictive relation between economicpolicy uncertainty and future stock market volatility in a multi-country analysis.

KW - Panel data, predictability, long-range dependence, Diebold-Mariano test, encompassing test

M3 - Working paper

T3 - CREATES Research Papers

BT - Assessing predictive accuracy in panel data models with long-range dependence

PB - Institut for Økonomi, Aarhus Universitet

CY - Aarhus

ER -