Errors in Repeated Measures Model function fitrm

Question

Minahil Khan 2018-2-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/382094-errors-in-repeated-measures-model-function-fitrm

关闭： MATLAB Answer Bot 2021-8-20

I'm trying to run a repeated measures ANOVA on my data in MATLAB. To do this I need to make my table a RepeatedMeasuresModel using fitrm. I've attached the .mat file with my table (tabn1.mat) Each participant in the trial had a value for Edge1, Edge2... until Center. Var1 is the participant's identification number. I set it as the predictor variable because I just want to do a comparison of means (compare average of everyone's edge1 values to edge2...etc)

Here's what I did:

Meas = table([1 2 3 4 5],'VariableNames',{'Measurements'});
tab1 = fitrm(tabn1, 'Edge1-Center~Var1', 'WithinDesign' Meas);

But then I get the following errors:

Error using eig
Input matrix contains NaN or Inf.
Error in RepeatedMeasuresModel>mauchlyTest (line 1922)
lam = eig(C'*S*C);
Error in RepeatedMeasuresModel.fit (line 1370)
            [this.Mauchly,this.Epsilon] = mauchlyTest(this.Cov,size(Xmat,1),rank(Xmat));
Error in fitrm (line 67)
s = RepeatedMeasuresModel.fit(ds,model,varargin{:});

I followed the example that Mathworks provides on their website very closely because I am not very familiar with statistics. I've also attached the example i followed.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再回答此问题。

Answer 1

Scott MacKenzie 2021-7-28

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/382094-errors-in-repeated-measures-model-function-fitrm#answer_755514

编辑：Scott MacKenzie 2021-7-28

testdata.txt

Here's an image of the data in this question:

The data are in the attached file, testdata.txt.

Design summary: This is a single-factor within-subjects design with six participants. The factor, or independent variable, is not named in the question, so it is given the name IV in the code below. There were six measures taken on the independent variable: Edge1, Edge2, Edge3, Edge4, and Center. The dependent variable is named Measure.

Below is the code to analyse the data using the ranova function. As seen in the anova table below, the effect of IV on Measure was not statistically significant: F(4,20) = 2.675, p > .05.

The code finishes with a function I wrote that generates a conventional anova table from the overly-complex and confusing anova table generated by the ranova function.

% data from question (1 within-subjects factor with 5 levels, 6 participants)
T = readtable('testdata.txt'); % 6x6
T.Properties.VariableNames = {'Participant' 'Edge1' 'Edge2' 'Edge3' 'Edge4' 'Center'};
% create the within-subjects design
withinDesign = table([1 2 3 4 5]','VariableNames',{'IV'});
withinDesign.IV = categorical(withinDesign.IV);
% create the repeated measure model and do the anova
rm = fitrm(T,'Edge1-Center ~ 1','WithinDesign',withinDesign);
AT = ranova(rm,'WithinModel','IV'); % remove comma to see ranova's table
% output a conventional anova table
disp(anovaTable(AT, 'Measure (units)'));
ANOVA table for Measure (units)
========================================================================
Effect            df          SS             MS         F         p
------------------------------------------------------------------------
Participant        5        0.00204        0.00041                      
IV                 4        0.00073        0.00018      2.675     0.0618
Participant(IV)   20        0.00136        0.00007                      
========================================================================
% ---------------------------------------------------------------------
% Scott's function to create a conventional ANOVA table from the
% overly-complicated and confusing anova table created by the ranova
% function.
function [s] = anovaTable(AT, dvName)
c = table2cell(AT);
% remove erroneous entries in F and p columns 
for i=1:size(c,1)       
        if c{i,4} == 1
            c(i,4) = {''};
        end
        if c{i,5} == .5
            c(i,5) = {''};
        end
end
% use conventional labels in Effect column
effect = AT.Properties.RowNames;
for i=1:length(effect)
    tmp = effect{i};
    tmp = erase(tmp, '(Intercept):');
    tmp = strrep(tmp, 'Error', 'Participant');
    effect(i) = {tmp}; 
end
% determine the required width of the table
fieldWidth1 = max(cellfun('length', effect)); % width of Effect column
fieldWidth2 = 57; % field needed for df, SS, MS, F, and p columns
barDouble = sprintf('%s\n', repmat('=', 1, fieldWidth1 + fieldWidth2));
barSingle = sprintf('%s\n', repmat('-', 1, fieldWidth1 + fieldWidth2));
% re-organize the data 
c = c(2:end,[2 1 3 4 5]);
c = [num2cell(repmat(fieldWidth1, size(c,1), 1)), effect(2:end), c]';
% create the ANOVA table
s = sprintf('ANOVA table for %s\n', dvName);
s = [s barDouble];
s = [s sprintf('%-*s %4s %11s %14s %9s %9s\n', fieldWidth1, 'Effect', 'df', 'SS', 'MS', 'F', 'p')];
s = [s barSingle];
s = [s, sprintf('%-*s %4d %14.5f %14.5f %10.3f %10.4f\n', c{:})];
s = [s, barDouble];
end