Continuous Record Asymptotics for Rolling Sample Variance Estimators.
It is widely known that conditional covariances of asset returns change over time. Researchers doing empirical work have adopted many strategies for accommodating conditional heteroskedasticity. One popular strategy is performing rolling regressions in which only data from, say, the preceding five year period is used to estimate the conditional covariance of returns at a given date. The authors develop continuous record asymptotic approximations for the measurement error in conditional variances when using these methods. They derive asymptotically optimal window lengths for the standard rolling regressions and optimal weights for weighted rolling regressions. The S&P 500 is used as an empirical example. Copyright 1996 by The Econometric Society.
Year of publication: |
1996
|
---|---|
Authors: | Foster, Dean P ; Nelson, Daniel B |
Published in: |
Econometrica. - Econometric Society. - Vol. 64.1996, 1, p. 139-74
|
Publisher: |
Econometric Society |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Asymptotic Filtering Theory for Univariate ARCH Models.
Nelson, Daniel B, (1994)
-
An Operational Measure of Riskiness
Foster, Dean P, (2009)
-
VIF Regression: A Fast Regression Algorithm for Large Data
Lin, Dongyu, (2011)
- More ...