repr_dwt {TSrepr}

DWT representation

Description

The repr_dwt computes DWT (Discrete Wavelet Transform) representation (coefficients) from a time series.

Usage

repr_dwt(x, level = 4, filter = "d4")

Arguments

x

the numeric vector (time series)

level

the level of DWT transformation (default is 4)

filter

the filter name (default is "d6"). Can be: "haar", "d4", "d6", ..., "d20", "la8", "la10", ..., "la20", "bl14", "bl18", "bl20", "c6", "c12", ..., "c30". See more info at wt.filter.

Details

This function extracts DWT coefficients. You can use various wavelet filters, see all of them here wt.filter. The number of extracted coefficients depends on the level selected. The final representation has length equal to floor(n / 2^level), where n is a length of original time series.

Value

the numeric vector of DWT coefficients

Author(s)

Peter Laurinec, <[e-mail]>

References

Laurinec P, Lucka M (2016) Comparison of representations of time series for clustering smart meter data. In: Lecture Notes in Engineering and Computer Science: Proceedings of The World Congress on Engineering and Computer Science 2016, pp 458-463

See Also

repr_dft, repr_dct, dwt

Examples

# Interpretation: DWT with Daubechies filter of length 4 and
# 3rd level of DWT coefficients extracted.
repr_dwt(rnorm(50), filter = "d4", level = 3)