PortfolioEffectEstim (version 1.4)

noise_acnv: Autocovariance Noise Variance

Description

Autocovariance Noise Variance (ACNV) estimates the noise variance based on the autocovariance of returns, rather than the Rescaled Noise Variance (RNV). It is generally preferred to RNV as it leads to a reduction in MSE and is robust to the presence of rare jumps. Also, this approach can be extended straightforwardly to estimate the parameters of higher order noise dependence.

Usage

noise_acnv(estimator)

Arguments

estimator

Vector of (time, price) observations for market asset when external market data is used.

Value

Details

- Convergence speed: $m^{1/2}$ (m - number of observation)

- Accounts for additive noise: yes

- Accounts for finite price jumps: yes

- Accounts for time dependence in noise: yes

- Accounts for endogenous effects in noise: no

References

R. C. Oomen, "Comment on realized variance and market microstructure noise by peter r. hansen and asger lunde," pp. 1-15, 23 September, 2005.

See Also

noise_rnv noise_urnv noise_uznv

Examples

## Not run: 
# data(spy.data) 
# estimator=estimator_create(priceData=spy.data)
# estimator_settings(estimator,
# 				   inputSamplingInterval = '10s',
# 				   resultsSamplingInterval = '10s')
# util_plot2d(noise_acnv(estimator),title="ACNV")
# ## End(Not run)