MATLAB CODE FOR ROOT MEAN SQUARED

Hello,please I have a matrix M, containng three distinct sets of data, and I intend to calculate the root mean squared. Coiuld some kinndly assist me; here is my trial code

clear all;close all;clc

A=[2 4 7 10 28]; %first
B=[1 3.6 8 11 27];%seccond
C=[0.5 5 9.5 13 32];%third
M=[A' B' C'];

[rows, columns] = size(M);
MEAN=zeros(rows, columns);
for row = 1 : rows
    MEAN(row) = mean(M(row, :));
    cal_rms=sqrt((MEAN-M(row, :)).^2./columns);
    
end
plot(cal_rms);

Answer 1

[Torsten]

This is not the standard definition for the root-mean-square value of a matrix.

Are you sure about the formula ? Then use

cal_rms(row) = sqrt((MEAN-M(row, :)).^2./columns);

instead of

cal_rms=sqrt((MEAN-M(row, :)).^2./columns);

[Grace]

Hi Torsten, It does not work, still woth

cal_rms(row) = sqrt((MEAN-M(row, :)).^2./columns);

[Torsten]

MATLAB has a function "rms" for the rooot-mean-square value - you might want to use it, but it doesn't work with the matrix elements shifted by the mean value.

What you calculate is some kind of standard deviation of the rows of M, but in this case, you had to divide by (columns-1), not by (columns).

A=[2 4 7 10 28]; %first
B=[1 3.6 8 11 27];%seccond
C=[0.5 5 9.5 13 32];%third
M=[A.', B.', C.'];
[rows, columns] = size(M);
MEAN=zeros(rows,1);
for row = 1 : rows
    MEAN(row) = mean(M(row, :));
    cal_rms(row) = sqrt(sum((MEAN(row)-M(row,:)).^2)/columns);    
end
plot(cal_rms)

Answer 2

[Bruno Luong]

To my book the mean should not be removed in calculation of RMS

M = rand(3,100);
[rows, columns] = size(M);
cal_rms = zeros(rows,1);
for row=1:size(M,1)
    cal_rms(row) = sqrt( sum(M(row, :).^2) / columns );
end
cal_rms % theoretical value is sqrt(1/3) for rand()

cal_rms = 3×1

0.5737 0.5854 0.5894