Why does matlab skip my if statement

In my MATLAB script, if statement does not function properly. I have found a way to avoid the part that is malfunctioning, but I wonder what went wrong. All if statements work, except for 'kg == 3.9'. The software does not recognize this and it skips this statement, and continues to process the data with the 'else' statement. In the image provided, you can see the problem while I was debugging. The script that is used is attached below.

clear;
kg = 0;
i = 1;
j = 1;
while kg < 6.6;
    filename = ['experiment2_', num2str(kg), 'kg.sig'];
    [data,extrainfo]=loadmat(filename);
    % First column is EMG data of Deltoid, second and third are from
    % contralateral trunk muscles.

      if kg == 2.6 || kg == 3.9
          data(:,1) = data(:,1)*2;
      elseif kg == 5.2 || kg == 6.5
          data(:,1) = data(:,1)*4;
          data(:,2:3) = data(:,2:3)*2;
      else 
          data = data;
      end

      % Time axis
      N=length(data);fs = 2000;dt = 1/fs;k=[0:N-1];t=k*dt;

      % ARV
      data_ARV = mean(abs(data));
      ARV(i,1:3)=data_ARV;

      % RMS
      data_RMS = sqrt(mean(data.^2));
      RMS(i,1:3)=data_RMS; 

      % Determine mean power frequency (MPF) and median frequency (MF)
      % data is the raw EMG, sfreq is the sample frequency
      [p,f]=pwelch(data,fs,round(0.9*fs),fs,fs);
      MPF(i,1:3)=sum(p.*f)./sum(p);
      x=sum(p);
      x1=cumsum(p);
      [m,ind]=min(abs(0.5*x-x1));
      MF(i,1:3)=f(ind);

      kg = kg+1.3;
      i = i+1;
  end

[Stephen23]

Sebastian Castro's answer tells you why, and what you can do about it. You also can read more here:

https://www.mathworks.com/matlabcentral/answers/57444-faq-why-is-0-3-0-2-0-1-not-equal-to-zero

http://matlab.wikia.com/wiki/FAQ#Why_is_0.3_-_0.2_-_0.1_.28or_similar.29_not_equal_to_zero.3F

https://www.mathworks.com/matlabcentral/answers/41536-why-is-2-24-10-22-4-not-equal-to-0

https://www.mathworks.com/matlabcentral/answers/321709-strange-behaviour-of-matlab-find-command

https://www.mathworks.com/matlabcentral/answers/329124-how-does-6-022140857-10-23-differ-from-6-022140857e-23

Answer 1

[Sebastian Castro]

This is a common issue when working with floating-point numbers, and I would first refer you to this answer: https://www.mathworks.com/matlabcentral/answers/57444-faq-why-is-0-3-0-2-0-1-not-equal-to-zero

You can easily check your specific case in MATLAB, since you're starting with a value of 2.6 and adding 1.3 to it. What does the following expression return?

(2.6 + 1.3) == 3.9

Typically, floating-point equality should be done within some tolerance -- for example

abs(kg - 3.9) < tol

where tol is a (usually) small number that makes sense for your application.

Sebastian