Plots the rolling correlations along with other required statistics to visualise the approach of Gershunov et al. (2001) to test the significance of signal from rolling correlation analysis.
Usage
rolCorPlot(x , y , width, start = 1, level = 0.95, main = NULL,
SDtest = TRUE, N = 500)
Arguments
x
A numeric vector.
y
A numeric vector.
width
A numeric vector of window lengths of the rolling correlation analysis.
start
The time of the first observation
level
Confidence level for intervals.
main
The main title of the plot.
SDtest
Set to TRUE to run test the significance of signal from rolling correlation analysis along with plotting.
N
An integer showing the number of series to be generated in Monte Carlo simulation.
Value
rolCor
A matrix showing rolling correlations for each width on its columns.
rolcCor.avr.filtered
A vector showing average rolling correlations filtered by running median nonlinear filter against outliers.
rolcCor.avr.raw
A vector showing unfiltered average rolling correlations.
rolCor.sd
A vector showing standard deviations of rolling correlations for each width.
rawCor
Pearson correlation between two series.
sdPercentiles
Percentiles of MC distribution of standard deviations of rolling correlations as the test limits.
test
A data frame showing the standard deviations of rolling correlations for each width along with level and (1-level) limits.
References
Gershunov, A., Scheider, N., Barnett, T. (2001). Low-Frequency Modulation of the ENSO-Indian Monsoon Rainfall Relationship: Signal or Noise? Journal of Climate, 14, 2486 - 2492.