qda_ews {earlywarnings}

Quick Detection Analysis for Generic Early Warning Signals

Description

Estimate autocorrelation, variance within rolling windows along a timeseries, test the significance of their trends, and reconstruct the potential landscape of the timeseries.

Usage

qda_ews(
  timeseries,
  param = NULL,
  winsize = 50,
  detrending = c("no", "gaussian", "linear", "first-diff"),
  bandwidth = NULL,
  boots = 100,
  s_level = 0.05,
  cutoff = 0.05,
  detection.threshold = 0.002,
  grid.size = 50,
  logtransform = FALSE,
  interpolate = FALSE
)

Arguments

timeseries

a numeric vector of the observed univariate timeseries values or a numeric matrix where the first column represents the time index and the second the observed timeseries values. Use vectors/matrices with headings.

param

values corresponding to observations in timeseries

winsize

is the size of the rolling window expressed as percentage of the timeseries length (must be numeric between 0 and 100). Default is 50%.

detrending

the timeseries can be detrended/filtered prior to analysis. There are four options: gaussian filtering, linear detrending and first-differencing. Default is no detrending.

bandwidth

is the bandwidth used for the Gaussian kernel when gaussian filtering is applied. It is expressed as percentage of the timeseries length (must be numeric between 0 and 100). Alternatively it can be given by the bandwidth selector bw.nrd0 (Default).

boots

the number of surrogate data to generate from fitting an ARMA(p,q) model. Default is 100.

s_level

significance level. Default is 0.05.

cutoff

the cutoff value to visualize the potential landscape

detection.threshold

detection threshold for potential minima

grid.size

grid size (for potential analysis)

logtransform

logical. If TRUE data are logtransformed prior to analysis as log(X+1). Default is FALSE.

interpolate

logical. If TRUE linear interpolation is applied to produce a timeseries of equal length as the original. Default is FALSE (assumes there are no gaps in the timeseries).

Value

qda_ews produces three plots. The first plot contains the original data, the detrending/filtering applied and the residuals (if selected), autocorrelation and variance. For each statistic trends are estimated by the nonparametric Kendall tau correlation. The second plot, returns a histogram of the distributions of the Kendall trend statistic for autocorrelation and variance estimated on the surrogated data. Vertical lines represent the level of significance, whereas the black dots the actual trend found in the time series. The third plot is the reconstructed potential landscape in 2D. In addition, the function returns a list containing the output from the respective functions generic_RShiny (indicators); surrogates_RShiny (trends); movpotential_ews (potential analysis)

Author(s)

Vasilis Dakos, Leo Lahti, March 1, 2013 [e-mail]

References

Dakos, V., et al (2012).'Methods for Detecting Early Warnings of Critical Transitions in Time Series Illustrated Using Simulated Ecological Data.' PLoS ONE 7(7): e41010. doi:10.1371/journal.pone.0041010

See Also

generic_ews; ddjnonparam_ews; bdstest_ews; sensitivity_ews; surrogates_ews; ch_ews; movpotential_ews; livpotential_ews;

Examples

data(foldbif)
out <- qda_ews(foldbif, param = NULL, winsize = 50, 
      	          detrending='gaussian', bandwidth=NULL, 
	  boots = 10, s_level = 0.05, cutoff=0.05, 
	  detection.threshold = 0.002, grid.size = 50, 
	  logtransform=FALSE, interpolate=FALSE)