analysisEEG/stat_statTestCollection.m at master · LightingResearchCenter/analysisEEG · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
% Calculates mean, SD, etc. here
function statOut = stat_statTestCollection(vecConc, x1, x2, colHeader, statOut, statFuncPath, shapWilk_pThr, bartlett_pThr, handles)

    % modify the code if you start feeding matrices here instead of vectors
    currDir = cd;

    % get the size of the input
    %     vecConc
    %     colHeader
    %     statOut
    vecConc_nonNan = vecConc(~isnan(vecConc));
    n = length(vecConc_nonNan); % length of non-NaN value vector
    sizeIn = size(vecConc);

    x1 = x1(~isnan(x1));
    x2 = x2(~isnan(x2));

    %% BASIC ONES

        % just add lines here and store it to the structure if you want to
        % calculate something more
        statOut.mean = nanmean(vecConc);
        statOut.median = nanmedian(vecConc);
        statOut.SD   = nanstd(vecConc);

        % We use mainly 3rd party implementations of the statistical tests
        cd(statFuncPath)
        vecConc_nonNan = vecConc(~isnan(vecConc));

    %% STATISTICAL TESTS

        %% ANOVA

            % If Data are normal (Gaussian) and deviations are homogeneous then you can do an ANOVA
            % --> significant if p < 0.05
            if statOut.shapWilk_W(1) < 1 && statOut.btestOut_P <= bartlett_pThr

                % The ANOVA test makes the following assumptions about the data in X:
                    % All sample populations are normally distributed.
                    % All sample populations have equal variance.
                    % All observations are mutually independent.

                % the Matlab Statistics Toolbox
                alpha  = handles.anova_pThreshold;
                X      = vecConc;
                [statOut.anovaP, statOut.anovaTable, statOut.anovaStats] = anova1(X, [], 'off');
                    % Notches in the boxplot provide a test of group medians
                    % (see boxplot) different from the F test for means in
                    % the ANOVA table             '
            else

                statOut.anovaP     = NaN;
                statOut.anovaTable = NaN;
                statOut.anovaStats = NaN;
            end


        %% t-Test

            alpha  = handles.student_pThreshold;
            [statOut.tTest_h, statOut.tTest_p, statOut.tTest_ci, statOut.tTest_stats] = ttest2(x1,x2,alpha);

            % statOut.tTest_h = [];
            % statOut.tTest_p = [];


        %% Kruskal-Wallis test

            X = vecConc;
            [statOut.kruskWallisP, statOut.kruskWallisTable, statOut.kruskWallisStats] = kruskalwallis(X, [], 'off');

        %% Wilcoxon-Mann-Whithney Test, unpaired sample

            % --> significant if p < 0.05

            % We use Giuseppe Cardillo's code for this, same as RANKSUM
            % in MATLAB,  % http://www.mathworks.com/matlabcentral/fileexchange/25830
            dispFullResultsFlag = 0;
            statOut.mww_Stats = mwwtest(x1,x2,dispFullResultsFlag);

            % In mwwtest the method is determined automatically, so we
            % use the same method for the ransum -function as well
            if strcmp(statOut.mww_Stats.method, 'Normal approximation') == 1
                handles.wilcox_method = 'approximate';
            else
                handles.wilcox_method = 'exact';
            end

            % RANKSUM (Matlab Statistics Toolbox)
            [statOut.rankSum_P, statOut.rankSum_H, statOut.rankSumStats] = ranksum(x1,x2,'alpha',handles.wilcox_pThreshold,'method',handles.wilcox_method);

                % check why NaNs are coming out


    cd(currDir)