How to use a loop for an array

My question is:

I have an array with values ranging from 0 to 200 something, and I need to find the average value for a set range of the data i.e. I need to find the average value for 0 to 1, the average value of the data for 1-2, and so on until it hits the end. I would like to save that information in a new matrix that I can look up later on.

Any help is appreciated.

[James Tursa]

What about the boundary points? E.g., if a value is exactly 1, do you want it to be part of the 0-1 average and also the 1-2 average? Or do you want these averages to be mutually exclusive?

[KK]

I suppose the range should be greater than the lower bound, less than or equal to the upper bound. I'll have to try out your method.

Answer 1

[Image Analyst]

If you have the Statistics and Machine Learning Toolbox, you can use grpstats(). It's a single line of code:

% Create sample data
r = 200 * rand(100000, 1); 
% Now find means in every integer range.
means = grpstats(r, floor(r))

[KK]

That may have worked like a charm, thank you very much. That was simple too!

[Image Analyst]

One point to note is that if there are absolutely no means in some range, then that element will be missing from means. So just be aware of that.

[KK]

Is there a way to save the values used for each mean in an array so that I may apply further operations to that information?

[Image Analyst]

In my code above, means is already an array. Do you mean like an array of those arrays? Like you want a 2D array or an array of structures?

Answer 2

[James Tursa]

Does something like this do what you want?

n = ceil(max(values));
x = bsxfun(@ge,values(:),0:n-1) & bsxfun(@le,values(:),1:n);
averages = sum(bsxfun(@times,values(:),x)) ./ sum(x);
averages(isnan(averages)) = 0;

[KK]

The ceil function is pretty cool. I like the idea of using the bsxfun, but how would I be able to calculate the average for the data with a value between 0 and 1, or 1 to 2 (and so on), using this function, for all the values in the array?