rc.kernel {realized} | R Documentation |
Realized covariance calculation using a kernel estimator.
rc.kernel(x, y, q, align.period = 1, adj = TRUE, type = 0, cts = TRUE, makeReturns = FALSE,...)
x |
RealizedObject or TimeSeries for S+ |
y |
RealizedObject or TimeSeries for S+ |
q |
Number of lags |
align.period |
Align the returns to this period first |
cts |
Create calendar time sampling if a non realizedObject is passed |
makeReturns |
Prices are passed make them into log returns |
adj |
T to use dof adjustment |
type |
0-11 or a character string |
... |
... |
The different types of kernels can be found using rKernel.available().
Kernel estimate of realized covariance.
Scott Payseur <spayseur@u.washington.edu>
Ole E. Barndorff-Nielsen, Peter Reinhard Hansen, Asger Lunde, and Neil Shephard. Regular and modified kernel-based estimators of integrated variance: The case with independent noise. Working Paper, 2004.
Michiel de Pooter, Martin Martens, and Dick van Dijk. Predicting the daily covariance matrix for sp100 stocks using intraday data - but which frequency to use? Working Paper, October 2005.
J. E. Griffen and R. C. A. Oomen. Covariance measurement in the presence of non-synchronous trading and market microstructure noise. Working Paper, June 27th, 2006.
rRealizedVariance
,rv.kernel
, rKernel.available
,rKernel
data(msft.real.cts) data(ge.real.cts) # kernel realized covariance for CTS aligned at one minute returns # rc.kernel(x = msft.real.cts[[1]], y = ge.real.cts[[1]], q=1, type="bartlett", align.period=60)