GARCH models for daily stock returns: Impact of estimation frequency on Value-at-Risk and Expected Shortfall forecasts

D. Ardia; L.F. Hoogerheide

doi:10.1016/j.econlet.2014.02.008

GARCH models for daily stock returns: Impact of estimation frequency on Value-at-Risk and Expected Shortfall forecasts

D. Ardia, L.F. Hoogerheide

Econometrics and Data Science

Research output: Contribution to Journal › Article › Academic › peer-review

Abstract

We analyze the impact of the estimation frequency-updating parameter estimates on a daily, weekly, monthly or quarterly basis-for commonly used GARCH models in a large-scale study, using more than twelve years (2000-2012) of daily returns for constituents of the S&P 500 index. We assess the implication for one-day ahead 95% and 99% Value-at-Risk (VaR) forecasts with the test for correct conditional coverage of Christoffersen (1998) and for Expected Shortfall (ES) forecasts with the block-bootstrap test of ES violations of Jalal and Rockinger (2008). Using the false discovery rate methodology of Storey (2002) to estimate the percentage of stocks for which the model yields correct VaR and ES forecasts, we conclude that there is no difference in performance between updating the parameter estimates of the GARCH equation at a daily or weekly frequency, whereas monthly or even quarterly updates are only marginally outperformed. © 2014 Elsevier B.V.

Original language	English
Pages (from-to)	187-190
Journal	Economics Letters
Volume	123
Issue number	2
DOIs	https://doi.org/10.1016/j.econlet.2014.02.008
Publication status	Published - 2014

Bibliographical note

PT: J; NR: 15; TC: 0; J9: ECON LETT; PG: 4; GA: AG7TZ; UT: WOS:000335623000018

Access to Document

10.1016/j.econlet.2014.02.008

Persistent URL (handle)

Copy link address

Cite this

@article{a7c5f31301ce4d31b247bbd34ab41936,

title = "GARCH models for daily stock returns: Impact of estimation frequency on Value-at-Risk and Expected Shortfall forecasts",

abstract = "We analyze the impact of the estimation frequency-updating parameter estimates on a daily, weekly, monthly or quarterly basis-for commonly used GARCH models in a large-scale study, using more than twelve years (2000-2012) of daily returns for constituents of the S&P 500 index. We assess the implication for one-day ahead 95% and 99% Value-at-Risk (VaR) forecasts with the test for correct conditional coverage of Christoffersen (1998) and for Expected Shortfall (ES) forecasts with the block-bootstrap test of ES violations of Jalal and Rockinger (2008). Using the false discovery rate methodology of Storey (2002) to estimate the percentage of stocks for which the model yields correct VaR and ES forecasts, we conclude that there is no difference in performance between updating the parameter estimates of the GARCH equation at a daily or weekly frequency, whereas monthly or even quarterly updates are only marginally outperformed. {\textcopyright} 2014 Elsevier B.V.",

author = "D. Ardia and L.F. Hoogerheide",

note = "PT: J; NR: 15; TC: 0; J9: ECON LETT; PG: 4; GA: AG7TZ; UT: WOS:000335623000018",

year = "2014",

doi = "10.1016/j.econlet.2014.02.008",

language = "English",

volume = "123",

pages = "187--190",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - GARCH models for daily stock returns: Impact of estimation frequency on Value-at-Risk and Expected Shortfall forecasts

AU - Ardia, D.

AU - Hoogerheide, L.F.

N1 - PT: J; NR: 15; TC: 0; J9: ECON LETT; PG: 4; GA: AG7TZ; UT: WOS:000335623000018

PY - 2014

Y1 - 2014

N2 - We analyze the impact of the estimation frequency-updating parameter estimates on a daily, weekly, monthly or quarterly basis-for commonly used GARCH models in a large-scale study, using more than twelve years (2000-2012) of daily returns for constituents of the S&P 500 index. We assess the implication for one-day ahead 95% and 99% Value-at-Risk (VaR) forecasts with the test for correct conditional coverage of Christoffersen (1998) and for Expected Shortfall (ES) forecasts with the block-bootstrap test of ES violations of Jalal and Rockinger (2008). Using the false discovery rate methodology of Storey (2002) to estimate the percentage of stocks for which the model yields correct VaR and ES forecasts, we conclude that there is no difference in performance between updating the parameter estimates of the GARCH equation at a daily or weekly frequency, whereas monthly or even quarterly updates are only marginally outperformed. © 2014 Elsevier B.V.

AB - We analyze the impact of the estimation frequency-updating parameter estimates on a daily, weekly, monthly or quarterly basis-for commonly used GARCH models in a large-scale study, using more than twelve years (2000-2012) of daily returns for constituents of the S&P 500 index. We assess the implication for one-day ahead 95% and 99% Value-at-Risk (VaR) forecasts with the test for correct conditional coverage of Christoffersen (1998) and for Expected Shortfall (ES) forecasts with the block-bootstrap test of ES violations of Jalal and Rockinger (2008). Using the false discovery rate methodology of Storey (2002) to estimate the percentage of stocks for which the model yields correct VaR and ES forecasts, we conclude that there is no difference in performance between updating the parameter estimates of the GARCH equation at a daily or weekly frequency, whereas monthly or even quarterly updates are only marginally outperformed. © 2014 Elsevier B.V.

U2 - 10.1016/j.econlet.2014.02.008

DO - 10.1016/j.econlet.2014.02.008

M3 - Article

SN - 0165-1765

VL - 123

SP - 187

EP - 190

JO - Economics Letters

JF - Economics Letters

IS - 2

ER -