stationarityTest: Stationarity test using the multitaper spectrum estimation...

In JieGroup/stationarity: Test the Stationarity of Time Series

Description

This function uses the multitaper method to construct hypothesis tests to test whether a given one dimensional is stationary. The test can be either two-way ANOVA or Rank-based test

Usage

stationarityTest(X, alpha = 0.05, method = "parametric")

Arguments

X	Vector of 1D time series to test, whose length is at least 40
alpha	Float of significance level (between 0 and 1, default to 0.05)
method	String of method type ('nonparametric' or 'parametric', default to 'parametric'), 'nonparametric' means two-way ANOVA test and 'parametric' means rank-based test

Value

Boolean (False means rejection of the null hypothesis that the time seres is stationary)

Examples

data(ppp)
dat <- ppp$japan
plot(dat)
stationarityTest(X=dat, alpha=0.01, method='parametric')
plot(diff(log(dat)))
stationarityTest(X=diff(log(dat)), alpha=0.01, method='parametric')

JieGroup/stationarity documentation built on June 3, 2019, 6:14 p.m.