Spearman correlation in Matlab!

Hey Matlab users,

If I have two series of data:

a = [1 4 6 3 4 6 7 8]; b [34 56 34 56 79 23 48 28];

and I want to perform a Spearman correlation what would be the proper command in MATLAB? I need both p-value and RHO.

Thanks for your help.

Mehdi

Answer 1

[Wayne King]

Hi, If you have the Statistics Toolbox.

a = [1 4 6 3 4 6 7 8]; 
b = [34 56 34 56 79 23 48 28];
[RHO,PVAL] = corr(a',b','Type','Spearman');

Answer 2

[Rithy Khouy]

a = [1 4 6 3 4 6 7 8];

b = [34 56 34 56 79 23 48 28];

[RHO,PVAL] = corr(a',b','Type','Spearman');

Answer 3

[DEVANSHI]

n = input("Enter the number of observations: ");
x = zeros(1, n);
y = zeros(1, n);
for i = 1:n x(i) = input("Enter the x: ");
    y(i) = input("Enter the y: ");
end 
[sx, idx_x] = sort(x);
rx = zeros(1, n);
AF_x = 0;
i = 1;
while i <= n j = i;
    while j <= n && sx(j) == sx(i) j = j + 1;
    end avg_rank = mean(i:j-1);
    for k = i:j-1 rx(idx_x(k)) = avg_rank;
    end
    t = j - i;
    if t > 1 AF_x = AF_x + (t^3 - t) / 12;
    end
    i = j;
end 
[sy, idx_y] = sort(y);
ry = zeros(1, n);
AF_y = 0;
i = 1;
while i <= n j = i;
    while j <= n && sy(j) == sy(i) j = j + 1;
    end avg_rank = mean(i:j-1);
    for k = i:j-1 ry(idx_y(k)) = avg_rank; 
    end
    t = j - i; if t > 1 AF_y = AF_y + (t^3 - t) / 12;
    end
    i = j; 
end d = rx - ry;
d2 = sum(d .^ 2);
spc = 1 - (6 * (d2 + AF_x + AF_y)) / (n * (n^2 - 1));
disp("Spearman's Rank Correlation Coefficient:");
disp(spc);



    for j=1:n
        if x(j)<x(i)
            yrank=yrank+1
        elseif y(j)==y(i)&& j==i
            xcount=xcount+1
        end