costat2.4.1 package

Time Series Costationarity Determination

Undo autoreflection action for an EWS object (wd stationary)

Perform bootstrap stationarity test for time series

Produces plots from output of findstysol that attempt to group differe...

Convert wavelet coefficients for two time-varying functions into two f...

Computes localized autocovariance and searches for costationary soluti...

Perform running mean smoothing of an EWS object

Extractor function for `csFSS`

object.

Given two time series find some time-varying linear combinations that ...

Form a particular linear combination of two time series and assess the...

Computes localized (wavelet) autocovariance function

Computes a Linear Combination Test Statistics

Plots solutions that are identified by findstysols

Compute the time-localized (unconditional) variance for a time series

Concatenate a set of solution results into one set

Plots results of a Bootstrap Test of Stationarity

Plot a `csBiFunction`

object

Plot a `csFSS`

object.

Produce plots from a `csFSSgr`

object.

Plot localized autocovariance (lacv) object.

Compute p-value for parametric Monte Carlo test and optionally plot te...

Print a `csBiFunction`

object.

Print a`csFSS`

object.

Print `csFSSgr`

object.

Print lacv class object

Combine two time series using a time-varying linear combination.

Summarize a `csBiFunction`

object.

Summarize a `csFSS`

object.

Summarize a `csFSSgr`

object.

Summarizes a lacv object

A test statistic for stationarity

Contains functions that can determine whether a time series is second-order stationary or not (and hence evidence for locally stationarity). Given two non-stationary series (i.e. locally stationary series) this package can then discover time-varying linear combinations that are second-order stationary. Cardinali, A. and Nason, G.P. (2013) <doi:10.18637/jss.v055.i01>.

Maintainer: Guy Nason License: GPL (>= 2) Last published: 2023-09-06