Skip to contents

Perform an automatically loess based smoothing of a time series

Source: R/loess_auto.R
loess_auto.Rd

Perform an automatically loess based smoothing of a time series. The local polynomial regression with automatic smoothing parameter selection is based on an optimisation using the 'aicc' bias-corrected 'AIC' criterion and the 'gcv' generalized cross-validation criterion.

Usage

loess_auto(
  time_series = NULL,
  genplot = FALSE,
  print_span = FALSE,
  keep_editable = FALSE
)

Arguments

time_series

Series to smooth: a two-column matrix or data frame with depth or time in column 1 and the proxy value (e.g. the period of a tracked cycle, as from completed_series) in column 2; rows with NA are dropped for the span selection. Required (Default=NULL).

genplot

Plot the original series (blue), the smoothed curve (black) and the smoothed curve plus and minus the absolute residual (red, dashed) (logical, TRUE or FALSE). Default=FALSE.

print_span

Print the selected loess span, a fraction (0-1) of the number of data points that is used in each local fit (logical, TRUE or FALSE). Default=FALSE.

keep_editable

Keep the graphical parameters (par) of the plot after the function returns, so that features can be added to it (logical, TRUE or FALSE); FALSE restores the previous settings. Default=FALSE.

Value

A matrix with 3 columns. The first column is depth/time. The second column is the smoothed curve. The third column is difference between the original curve and the smoothed curve.

References

Cleveland, W. S. (1979) Robust locally weighted regression and smoothing scatter plots. Journal of the American Statistical Association. 74, 829–836. <doi:10.1080/01621459.1979.10481038> Hurvich, C.M., Simonoff, J.S., and Tsai, C.L. (1998), Smoothing Parameter Selection in Nonparametric Regression Using an Improved Akaike Information Criterion. Journal of the Royal Statistical Society B. 60, 271–293 <doi:10.1111/1467-9868.00125> Golub, G., Heath, M. and Wahba, G. (1979). Generalized cross validation as a method for choosing a good ridge parameter. Technometrics. 21, 215–224. <doi:10.2307/1268518>

Author

Based on the the loess.as function of the 'fANCOVA' R package.

Examples

# \donttest{
#'smooth the period curve of the 405 kyr eccentricity cycle extracted from
# the magnetic susceptibility data set of Pas et al., (2018)
#perform the CWT on the magnetic susceptibility data set of Pas et al., (2018)

mag_wt <- analyse_wavelet(data = mag,
dj = 1/100,
lowerPeriod = 0.1,
upperPeriod = 254,
verbose = FALSE,
omega_nr = 10)

#Track the 405 kyr eccentricity cycle in a wavelet spectra

#mag_track <- track_period_wavelet(astro_cycle = 405,
#                                   wavelet=mag_wt,
#                                   n.levels = 100,
#                                   periodlab = "Period (metres)",
#                                   x_lab = "depth (metres)")

#Instead of tracking, the tracked solution data set mag_track_solution is used
mag_track <- mag_track_solution

mag_track_complete <- completed_series(
  wavelet = mag_wt,
  tracked_curve = mag_track,
  period_up = 1.2,
  period_down = 0.8,
  extrapolate = TRUE,
  genplot = FALSE,
  keep_editable=FALSE
)

#Smooth the completed tracking of the 405 kyr eccentricity cycle as tracked in the wavelet spectra
mag_track_complete <- loess_auto(time_series = mag_track_complete,
genplot = FALSE, print_span = FALSE,keep_editable=FALSE)
# }