spec_estimation {mvLSWimpute}

Function to estimate the Local Wavelet Spectral matrix for a multivariate locally stationary time series containing missing values

Description

This function estimates the LWS matrix for a multivariate locally stationary time series containing missing values. If the input time series does not contain missing values then spectral estimation is carried out using routines from the mvLSW package.

Usage

spec_estimation(data, interp = "linear")

Arguments

data

Input multivariate time series, matrix of dimension TxP where P is the number of channels and T is the length of the series.

interp

Method of interpolation of NAs in spectrum. Can be "linear" or "spline"; see na_interpolation for more detals. See also note below.

Value

Returns a mvLSW object containing the estimated LWS matrix.

Note

For some series with a lot of missing values, the linear interpolation will result in zero periodogram values (due to the form of the Haar filters). This may not be desirable, so a higher order (spline) interpolation function may be better.

See Also

correct_per, smooth_per, mvEWS, na_interpolation

Examples

## Sample bivariate locally stationary time series

set.seed(1)
X <- matrix(rnorm(2 * 2^8), ncol = 2)
X[1:2^7, 2] <- 3 * (X[1:2^7, 2] + 0.95 * X[1:2^7, 1])
X[-(1:2^7), 2] <- X[-(1:2^7), 2] - 0.95 * X[-(1:2^7), 1]
X[-(1:2^7), 1] <- X[-(1:2^7), 1] * 4
X <- as.ts(X)

# create some missing values, taking care to provide some data at the start of the series 

missing.index = sort(sample(10:2^8, 30))

X[missing.index, ] <-NA

# estimate the spectrum

spec = spec_estimation(X)