function [logRS,logERS,V]=RSana(x,n,method,q)
%Syntax: [logRS,logERS,V]=RSana(x,n,method,q)
%____________________________________________
%
% Performs R/S analysis on a time series.
%
% logRS is the log(R/S).
% logERS is the Expectation of log(R/S).
% V is the V statistic.
% x is the time series.
% n is the vector with the sub-periods.
% method can take one of the following values
% (1)'Hurst' for the Hurst-Mandelbrot variation.
% (2)'Lo' for the Lo variation.
% (3)'MW' for the Moody-Wu variation.
% (4)'Parzen' for the Parzen variation.
% q can be either
% (1)a (non-negative) integer.
% (2)'auto' for the Lo's suggested value.
%
%
% References:
%
% Peters E (1991): Chaos and Order in the Capital Markets. Willey
%
% Peters E (1996): Fractal Market Analysis. Wiley
%
% Lo A (1991): Long term memory in stock market prices. Econometrica
% 59: 1279-1313
%
% Moody J, Wu L (1996): Improved estimates for Rescaled Range and Hurst
% exponents. Neural Networks in Financial Engineering, eds. Refenes A-P
% Abu-Mustafa Y, Moody J, Weigend A: 537-553, Word Scientific
%
% Hauser M (1997): Semiparametric and nonparametric testing for long
% memory: A Monte Carlo study. Empirical Economics 22: 247-271
%
%
% Alexandros Leontitsis
% Department of Education
% University of Ioannina
% 45110 - Dourouti
% Ioannina
% Greece
%
% University e-mail: me00743@cc.uoi.gr
% Lifetime e-mail: leoaleq@yahoo.com
% Homepage: http://www.geocities.com/CapeCanaveral/Lab/1421
%
% 1 Jan 2004.
% % method=Hurst;
if nargin<1 | isempty(x)==1
error('You should provide a time series.');
else
% x must be a vector
if min(size(x))>1
error('Invalid time series.');
end
x=x(:);
% N is the time series length
- 1
- 2
前往页