Rank: 370959 based on 0 downloads (last 30 days) and 0 file submitted
photo

Natalia

E-mail

Personal Profile:
Professional Interests:

 

Watch this Author's files

 

Comments and Ratings by Natalia View all
Updated File Comments Rating
16 May 2012 Circular Statistics Toolbox (Directional Statistics) Compute descriptive and inferential statistics for circular or directional data. Author: Philipp Berens

I confirm Dillon's report on circ_wwtest bug.

01 Mar 2012 Circular Statistics Toolbox (Directional Statistics) Compute descriptive and inferential statistics for circular or directional data. Author: Philipp Berens

Thank you very much for such a useful toolbox. Now, I have a question related to circ_ktest (two-smple test to compare concentration). The F statistic is defined only in case of rbar>.7, Mardia (pag 133, 1999) compute F in the case where resultant vector length is <0.45 :
n1 = length(alpha1);
n2 = length(alpha2);

R1avg=circ_r(alpha1);
R2avg = circ_r(alpha2);

R1 = n1*circ_r(alpha1);
R2 = n2*circ_r(alpha2);

%make sure that rbar > .7
rbar = (R1+R2)/(n1+n2);

if rbar > .7

f = ((n2-1)*(n1-R1))/((n1-1)*(n2-R2));

elseif rbar< .45 %taken from Mardia 1999 p.133 (Baschelet report: Mardia 1972 pag 161)

g11= asin(2*sqrt(3/8)*(R1avg));
g12= asin(2*sqrt(3/8)*(R2avg));

f= (2/sqrt(3))*((g11-g12)/(1/(n1-4)+ 1/(n2-4)).^(1/2));

But here Sample 1 and Sample 2 define the sign of F... and so S1 and S2 will be defined depending on Ravg value being S1>S2 for computation of F. Is this right?
Thank you!
natalia

Contact us