MATLAB: Averaging Data in Buckets of Size n

MATLAB

I have 250000 data points and I would like to filter that such that every 10 data points is averaged to one data point. How can I do this?

Best Answer

You can reshape the matrix such that it has 'n' rows. Once this is done, you can take the mean of the reshaped matrix and this will give the mean of every 'n' elements. Try the following:

myData = 1:200;
n = 10;
% Reshape the matrix such that it has 'n' rows
reshapedData = reshape(myData, n, [ ]);
% Take the average of each column
averageMatrix = mean(reshapedData);

Related Solutions

MATLAB: How to group and average non-sequential vectors

This workflow can be easily achieved when working with a table:

1. Using a sample data:

>> n = 500;
>> time = (1:n)';
>> X = 45*rand(n,1);
>> Y = 100*rand(n,1) + 10;

2. Store your data within a table:

>> T = table(time, X, Y);

3. Create a new categorical variable in the table which identify the grouping for each row using "discretize":

>> edges = (0:5:45)';
>> T.myGroup = discretize(T.X, edges,'categorical');

4. Now, you can get the mean of each group using "grpstats":

>> meanTable = grpstats(T, 'myGroup', 'mean');
>> meanTable.mean_time = []

5. and plot the mean:

>> figure
>> plot(meanTable.mean_X, meanTable.mean_Y)

MATLAB: Bucketing values

If you have the Statistics Toolbox installed, there is the built-in RANDSAMPLE command to do this.

A = [0.8756 0.1185 0.0059];
X = randsample(numel(A),5000,true, A);  % X gets 5000 samples
tabulate(X)

Best Answer

Related Solutions

MATLAB: How to group and average non-sequential vectors

MATLAB: Bucketing values

Related Question