quant_test.zip

Data normalization of an 2 dimensional numerical array

% option={'median','mean','trimean','geomean','harmean'};
%
% How it works?
% Qunatile normalization is simple and effective
% 1 Step: sort all columns amd store the sorted data and the sort index
% 2 Step: calculate the row 'median','mean','trimean','geomean','harmean'
% of the sorted data
% 3 Step: replace the original raw value with the matching quantile value
%
% ExExample with and 6 x 4 array
% Input array
%
% 668 1515 469 772
% 1148 1631 1500 593
% 439 931 449 441
% 1473 236 1711 1041
% 500 254 645 733
% 992 438 362 689
% Step 1: sorted raw data
% 439 236 362 441
% 500 254 449 593
% 668 438 469 689
% 992 931 645 733
% 1148 1515 1500 772
% 1473 1631 1711 1041

% Step 2: median value for each quantile
% 400.5
% 474.5
% 568.5
% 832
% 1324
% 1552

% Step 3: replace the original raw data value with the matching quantile
% value. This is done with the resort index
% 3 5 3 5 568.5 1324 568.5 1324
% 5 6 5 2 1324 1552 1324 474.5
% 1 4 2 1 400.5 832 474.5 400.5
% 6 1 6 6 1552 400.5 1552 1552
% 2 2 4 4 474.5 474.5 832 832
% 4 3 1 3 832 568.5 400.5 568.5

% Job done