Compute the standard error accounting for empirical autocorrelations
seCor(
x,
effCor = if (missing(effCov)) computeEffectiveAutoCorr(x) else effCov/var(x, na.rm =
TRUE),
na.rm = FALSE,
effCov,
nEff = computeEffectiveNumObs(x, effCor, na.rm = na.rm)
)numeric vector
numeric vector of effective correlation components
first entry at zero lag equals one. See computeEffectiveAutoCorr
logical. Should missing values be removed?
alternative to specifying effCor: numeric vector of
effective covariance components
first entry is the variance. See computeEffectiveAutoCorr
possibility to specify precomputed number of effective observations for speedup.
numeric scalar of standard error of the mean of x
The default uses empirical autocorrelation
estimates from the supplied data up to first negative component.
For short series of x it is strongly recommended to to
provide effCov that was estimated on a longer time series.