How to isolate particular values from matrix containing Spearman rank correlation coefficient

1 次查看（过去 30 天）

Torrey Beek 2013-6-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/79938-how-to-isolate-particular-values-from-matrix-containing-spearman-rank-correlation-coefficient

Hi,

I hope I can manage to ask this problem clearly!

I am using a data set that contains two matrices both (365x48) in dimension. I first perform a smoothing function using a 29 day window to smooth both data sets, as seen in the code below:

for i=1:365;
    p=1:48;
        if i<=14
            KILB11_AP_SW_PU_Matrix(i,p)=mean(KILB11_AP_PU_Matrix([(i-1)+352:365 1:i+14],p));
        else if i>=352
            KILB11_AP_SW_PU_Matrix(i,p)=mean(KILB11_AP_PU_Matrix([i-14:365 1:1+(i-352)],p));        
        else KILB11_AP_SW_PU_Matrix(i,p)=mean(KILB11_AP_PU_Matrix((i-14):(i+14),p));
        end
end
end

What I would like to do is to analyse how the value of the Spearman rank correlation coefficient changes throughout the year between my two variables. Ideally I would compute the value of the coefficient for each specific element in both matrices by using the same 29 day window principle to calculate the coefficient.

I am hopeful that there may be someway to isolate the value of the coefficient from the resultant matrix output from using the 'corr(MATRIX,'type','Spearman').

Any ideas?