plot_wave.local.multiple.cross.correlation {wavemulcor}

Auxiliary routine for plotting wave local multiple cross-correlations

Description

Produces a plot of wave local multiple cross-correlations.

Usage

plot_wave.local.multiple.cross.correlation(Lst, lmax,
     lag.first=FALSE, xaxt="s", pdf.write=NULL)

Arguments

Lst

A list from local.multiple.cross.regression.

lmax

maximum lag (and lead).

lag.first

if TRUE, it produces lag-lead pages with J+1 wavelet plots each. Otherwise (default) it gives wavelet pages with 2*lmax+1 lag-lead plots each.

xaxt

An optional vector of labels for the "x" axis. Default is 1:n.

pdf.write

Optional name leader to save files to pdf format. The actual filename of each page is either "plot_<pdf.write>_WLMCC_<i>.pdf", where "i" is the lag/lead i=-lmax...+lmax, or, "plot_<pdf.write>_WLMCC_<j>.pdf", where "j" is the wavelet level j=1...(J+1).

Details

The routine produces time series plots of wave local multiple cross-correlations with their confidence intervals. Also, at every upturn and downturn, the name of the variable that maximizes its multiple cross-correlation against the rest is shown.

Value

Plot.

Author(s)

Javier Fernández-Macho, Dpt. of Quantitative Methods, University of the Basque Country, Agirre Lehendakari etorb. 83, E48015 BILBAO, Spain. (email: javier.fernandezmacho at ehu.eus).

References

Fernández-Macho, J., 2018. Time-localized wavelet multiple regression and correlation, Physica A: Statistical Mechanics, vol. 490, p. 1226–1238. <DOI:10.1016/j.physa.2017.11.050>