DCCA.dcca
— Methoddcca(x,y; box_start = 3, box_stop = div(length(x),10), nb_pts = 30)
Performs the DCCA analysis of x
and y
. The default analysis starts with a window size of 3 up to one tenth of the total length of x
for statistical reasons.
returns the dcca coefficients.
DCCA.empirical_CI
— Methodempirical_CI(x,y) --> points, critical_values
Extrapolates tables found in the litterature to provide a 95% confidence interval between start
and stop
. The confidence interval represents the null hypothesis "no correlations". Provides "ready to plot" confidence interval, by also returning the points at which the critical values are evaluted.
DCCA.log_space
— Methodlog_space(start, stop, num)
Returns a linear spacing in loglog scale. The parameter start
and stop
define the range and num
corresponds to the number of desired points.
DCCA.rhoDCCA
— MethodrhoDCCA(x,y; box_start = 3, box_stop = div(length(x),10), nb_pts = 30)
Performs the DCCA analysis of x
and y
. The default analysis starts with a window size of 3 up to one tenth of the total length of x
for statistical reasons. returns the different window sizes used for the analysis, and the associated dcca coefficients.
DCCA.bootsrap_CI
— Methodbootstrap_CI(x,y) --> points, critical_values
Provides absolute bounds for the null hypothesis "no correlations" using a bootstrap procedure, as well as the time scales they are associated to. Provides "ready to plot" confidence interval.
DCCA.complete_CI
— Methodcomplete_CI(data_length, points)
returns the complete interpolated 95% confidence intervals for all the points in points
.
DCCA.detrending1
— Methoddetrending(x; order = 1)
Performs a linear detrending of x
. You can change the order of the polynomials to a higher order for a non-linear detrending.
DCCA.interpolate
— Methodinterpolate(x,y) -> interpolation_function
creates and return an interpolation function working for any point.
DCCA.nearest_neighbour
— Methodnearest_neighbour(value, listOfValues) --> (nn, index)
finds and returns the nearest neighbour to value
in listOfValues
, as well as the index of it in listOfValue
.
DCCA.partitioning
— Methodpartitioning(x, box_size; overlap = div(box_size,2))
Partitions x
into several segment of length box_size
. The default behavior is that the windowed data have a 50% overlap. To change this, change the overlap
parameter.
returns an array of size (n,box_size), n being the total number of segments.