GMZTests (version 0.1.4)

dfa.test: Statistical test for Detrended Fluctuation Analysis.

Description

This function performs the statistical test for the long-range correlation exponents obtained by the Detrended Fluctuation Analysis method.

Usage

dfa.test(y, npoints, rep, ts.sim, prob)

Arguments

y

A vector contaning univariate time series.

npoints

The number of different window sizes that will be used to estimate the Fluctuation function in each zone. See nonlinearTseries package.

rep

An integer value indicating the number of repetitions.

ts.sim

An logical value. If TRUE, the confidence interval for alpha_dfa is obtained from a White Gaussian Noise. If FALSE, the confidence interval for alpha_dfa is obtained from the shuffling of the original series.

prob

An numeric value indicating the quantile of probability to be used in estimating confidence intervals by N(0,1).

Value

An rbind matrix containing "alpha_dfa","se_alpha_dfa", "r2_alpha_dfa","min_alpha_dfa","max_test","mean_test", "median_test", "sd_test", "skewness_test", "kurtosis_test", "jarquebera_test_pvalue", and confidence interval: "CI_lower_test", "CI_upper_test".

Details

This function include following measures alpha_dfa, se_alpha_dfa, r2_alpha_dfa, min_test, max_test, mean_test, median_test, sd_test, skewness_test, kurtosis_test, Jarquebera_test_pvalue, CL_lower_test, CL_upper_test

References

KRISTOUFEK, L. Rescaled Range Analysis and Detrended Fluctuation Analysis: Finite Sample Properties and Confidence Intervals. AUCO Czech Economic Review, v.4,n.3, p.315-329, 2010.

Examples

# NOT RUN {
y=rnorm(1000)
dfa.test(y, npoints=15, rep=10,ts.sim="TRUE", prob=.95)

# }