Hi Daniela,
I understand that you are trying to average the data given in (x,y,z) coordinates over ‘n’ runs, where each run is having different data points due to missing data.
To solve this, you can use the interpolation method which generates missing data points. MATLAB provides several methods to interpolate, linear interpolation being one of them. After interpolating you can take the average over the ‘n’ runs. Follow the below steps to interpolate the data and calculate the average:
- Find maximum number of data points in a run by using: “max(cellfun(@(x) size(x, 1), data))”
- Create an empty matrix to store the interpolated values: “zeros(maxPoints, 3);”
- Determine number of data points using: ”size(data{i}, 1);”
- Linearly interpolate missing data using: “interp1(1:numPoints, data{i}, linspace(1, numPoints, maxPoints));”
- Store the data and calculate the average
An example code for the above steps is given below:
% Example data
n = 3; % Number of runs
data = cell(n, 1);
data{1} = [1 2 3; 4 5 6; 7 8 9]; % Run 1
data{2} = [10 11 12; 13 14 15]; % Run 2
data{3} = [16 17 18; 19 20 21; 22 23 24; 25 26 27]; % Run 3
% Step 1: Identify the maximum number of data points
maxPoints = max(cellfun(@(x) size(x, 1), data));
% Step 2: Loop through each run
for i = 1:n
% Step 2: Generate empty matrix
averagedData = zeros(maxPoints, 3);
% Step 3b: Determine the number of data points
numPoints = size(data{i}, 1);
% Step 3c: Interpolate missing data points
interpolatedData = interp1(1:numPoints, data{i}, linspace(1, numPoints, maxPoints));
% Step 3c: Store interpolated data
averagedData(1:maxPoints, :) = averagedData(1:maxPoints, :) + interpolatedData;
data{i} = averagedData;
disp(data{i})
end
% Step 4: Calculate the average
dim = ndims(data{1});
M = cat(dim+1,data{:});
meanArray = mean(M,dim+1);
disp(meanArray)
For more information on the “interp1” function, please refer to the following link:
Hope this helps!
Regards,
Ayush.
