2021-08-31 Version 3.1.2: Some reported bugs and vignette compilation errors corrected.
2020-05-18 Version 3.1.1: New heatmap_ routines included.
2020-03-05 Version 3.1.0: New plot_ routines included.
2020-02-01 Version 3.0.0: New .regression routines incorporate wavelet (and wave-local) regressions.
        That is, coefficients and c.i.'s, standard deviations, t-ratios and p-values are reported
        for all wavelet scales, time lags (.cross.regression) and over time (local. routines).
        They extend all previous routines; in other words, they also calculate wavelet (and wave-local)
        correlations as before.
        Nevertheless, previous .correlation routines are also kept for back compatibility.
2020-01-20 Version 2.2.3: wave.multiple.cross.correlation: c.i.'s calculation added;
        new argument p added (with default=.975).
        wave.multiple.correlation: Set argument N's default as unevaluated promise.
        It is used only to calculate c.i.'s df and actually not needed as it can be obtained
        from length(xx[[1]][[1]]). (Note it is not actually dropped for back compatibility).
2020-01-20 Version 2.2.2: Bivariate case: allow for negative correlations.
        Multivariate case: correct confidence interval: truncate to nonnegative values.
2018-06-08 Version 2.2.1 released.
        Version 2.2 incorporates an auxiliary function local.multiple.correlation.R
        that calculates local multiple correlations (LMC).
        They are calculated by letting move along time
        a window of weighted time series values around t. Any of the six weight
        functions mentioned in previous releases can be used.
