Compute mean for multiple different length data

Hello, I’ve been searching a solution for a while but to little avail. I have many data files with different sizes saved as .mat . I need to load them in and compute their means at the same X value. The file looks alike:

file1: -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 -- X value 0.17 0.25 0.14 0.54 0.5 0.34 0.11 0.33 0.91 1.0 0.72 0.65 0.83 0.32 -- data

file2: -1 0 1 2 3 4 -- X value 0.85 0.37 0.41 0.58 1.0 0.73 -- Data

file3, file4, etc…

I have a ‘ for’ loop to load file and get X and data, how to program them to have the same length and compute mean? I would like to patch the missing data point with NaN. The X value is increment of 1. I need your kind suggestions. Thank you!

[Oleg Komarov]

Vertical means?

[Yonghe]

Hi Oleg,

I need the Data mean value at each of X vale for all the files. Such as, in my example, the mean Data value at X = -5, -4, -3, -2, -1, 0, 1, ...

Answer 1

[Jan]

You can use 2 loops: One for reading and the 2nd for the computations:

XList    = zeros(2, nFile);
DataList = cell(1, nFile);
for iFile = 1:nFile
   % Get X and Data from file:
   ...
   XList(1:2, iFile) = [X(1); X(end)];
   DataList{iFile} = Data;
end

Xmin = min(XList(1, :));
Xmax = max(XList(2, :));
Xnum = Xmax - Xmin + 1;
DataSum = zeros(1, Xnum);
DataNum = zeros(1, Xnum);
for iFile = 1:nFile
  ini = XList(1, iFile) - Xmin + 1;
  fin = XList(2, iFile) - Xmin + 1;
  DataSum(ini:fin) = DataSum(ini:fin) + DataList{iFile};
  DataNum(ini:fin) = DataNum(ini:fin) + 1;
end

DataMean = DataSum ./ DataNum;

[Yonghe]

Jan & Fangjun, Thank you for your help! -- Yonghe

Answer 2

[Fangjun Jiang]

To fill the missing data with nan, use the following code. But once you fill it with nan, you won't be able to calculate a meaningful mean value, so you need to decide before proceeding to the next step.

x=[-1 0 1 2 3 4];
data=[0.85 0.37 0.41 0.58 1.0 0.73];

All_X=-10:10;
NewData=nan(size(All_X));
Index=ismember(All_X,x);
NewData(Index)=data;