Home > Research > Publications & Outputs > An Investigation into Multivariate Variance Rat...
View graph of relations

An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability. / Hong, Seok Young; Linton, Oliver; Zhang, Hui Jun .
In: Journal of Financial Econometrics, Vol. 15, No. 2, 21.03.2017, p. 173-222.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Hong SY, Linton O, Zhang HJ. An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability. Journal of Financial Econometrics. 2017 Mar 21;15(2):173-222. doi: 10.1093/jjfinec/nbw014

Author

Hong, Seok Young ; Linton, Oliver ; Zhang, Hui Jun . / An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability. In: Journal of Financial Econometrics. 2017 ; Vol. 15, No. 2. pp. 173-222.

Bibtex

@article{5d1fdd9e60d34869be666db55e25e482,
title = "An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability",
abstract = "We propose several multivariate variance ratio statistics for “testing” the weak form Efficient Market Hypothesis and for measuring the direction and magnitude of departures from this hypothesis. We derive the asymptotic distribution of the statistics and scalar functions thereof under the null hypothesis that returns are unpredictable after a constant mean adjustment. We propose asymptotic standard errors that are robust to departures from the “no leverage” assumption of Lo and MacKinlay (1988), but are relatively simple and in particular do not require the selection of a bandwidth parameter. We show the limiting behavior of the statistic under a multivariate fads model and under a moderately explosive bubble process: these alternative hypotheses give opposite predictions with regards to the long-run value of the statistics. We apply the methodology to weekly returns for Center for Research in Security Prices size-sorted portfolios from 1962 to 2013 in three subperiods. We find evidence of a reduction of linear predictability in the most recent period, for small and medium cap stocks, but we still reject the multivariate null hypothesis in the most recent period. The main findings are not substantially affected by allowing for a common factor time varying risk premium.",
author = "Hong, {Seok Young} and Oliver Linton and Zhang, {Hui Jun}",
year = "2017",
month = mar,
day = "21",
doi = "10.1093/jjfinec/nbw014",
language = "English",
volume = "15",
pages = "173--222",
journal = "Journal of Financial Econometrics",
issn = "1479-8409",
publisher = "Oxford University Press",
number = "2",

}

RIS

TY - JOUR

T1 - An Investigation into Multivariate Variance Ratio Statistics and their Application to Stock Market Predictability

AU - Hong, Seok Young

AU - Linton, Oliver

AU - Zhang, Hui Jun

PY - 2017/3/21

Y1 - 2017/3/21

N2 - We propose several multivariate variance ratio statistics for “testing” the weak form Efficient Market Hypothesis and for measuring the direction and magnitude of departures from this hypothesis. We derive the asymptotic distribution of the statistics and scalar functions thereof under the null hypothesis that returns are unpredictable after a constant mean adjustment. We propose asymptotic standard errors that are robust to departures from the “no leverage” assumption of Lo and MacKinlay (1988), but are relatively simple and in particular do not require the selection of a bandwidth parameter. We show the limiting behavior of the statistic under a multivariate fads model and under a moderately explosive bubble process: these alternative hypotheses give opposite predictions with regards to the long-run value of the statistics. We apply the methodology to weekly returns for Center for Research in Security Prices size-sorted portfolios from 1962 to 2013 in three subperiods. We find evidence of a reduction of linear predictability in the most recent period, for small and medium cap stocks, but we still reject the multivariate null hypothesis in the most recent period. The main findings are not substantially affected by allowing for a common factor time varying risk premium.

AB - We propose several multivariate variance ratio statistics for “testing” the weak form Efficient Market Hypothesis and for measuring the direction and magnitude of departures from this hypothesis. We derive the asymptotic distribution of the statistics and scalar functions thereof under the null hypothesis that returns are unpredictable after a constant mean adjustment. We propose asymptotic standard errors that are robust to departures from the “no leverage” assumption of Lo and MacKinlay (1988), but are relatively simple and in particular do not require the selection of a bandwidth parameter. We show the limiting behavior of the statistic under a multivariate fads model and under a moderately explosive bubble process: these alternative hypotheses give opposite predictions with regards to the long-run value of the statistics. We apply the methodology to weekly returns for Center for Research in Security Prices size-sorted portfolios from 1962 to 2013 in three subperiods. We find evidence of a reduction of linear predictability in the most recent period, for small and medium cap stocks, but we still reject the multivariate null hypothesis in the most recent period. The main findings are not substantially affected by allowing for a common factor time varying risk premium.

U2 - 10.1093/jjfinec/nbw014

DO - 10.1093/jjfinec/nbw014

M3 - Journal article

VL - 15

SP - 173

EP - 222

JO - Journal of Financial Econometrics

JF - Journal of Financial Econometrics

SN - 1479-8409

IS - 2

ER -