Error using accumarray: First input SUBS must contain positive integer subscripts?

Hi all,

I have tried Matlab function ranova but get an error:

Error using accumarray First input SUBS must contain positive integer subscripts.

Error in RepeatedMeasuresModel>makeTestA (line 2172) termsize = accumarray(this.CoefTerms',1);

Error in RepeatedMeasuresModel/ranova (line 659) [A,Anames] = makeTestA(this);

Could someone please let me know how to solve this problem?

Here is how I have tried ranova:

t = table(subjects,means(:,1),means(:,2),means(:,3),'VariableNames',{'sessions','s1','s2','s3'});

t =

    sessions      s1       s2        s3 
    _________    ______    ______    ______

    'subject'    5.0463    5.3267    5.5495
    'subject'    6.0667     7.087     3.322
    'subject'    3.1313    7.4503    3.2299
    'subject'    16.412    6.1687    5.0404
    'subject'     8.824    14.683    6.7319
    'subject'    3.8126    2.8653    3.6366
    'subject'     27.42    9.1804    8.0514
    'subject'    7.0583    4.9728    4.2714
    'subject'    3.7724    5.1526     5.678
    'subject'    12.036    4.4427    14.088
    'subject'    8.5065    5.4886    3.7804
    'subject'      6.64    10.418    6.3139
    'subject'    5.1549    19.468    9.2322
    'subject'    4.6753    4.4866    5.7057
    'subject'    3.3991     3.551    3.6003
    'subject'    10.692    11.041    4.9977

rm = fitrm(t,'s1-s3~sessions') rm =

RepeatedMeasuresModel with properties:

   Between Subjects:
         BetweenDesign: [16x4 table]
         ResponseNames: {'s1'  's2'  's3'}
    BetweenFactorNames: {'sessions'}
          BetweenModel: '1 + sessions'

   Within Subjects:
         WithinDesign: [3x1 table]
    WithinFactorNames: {'Time'}
          WithinModel: 'separatemeans'

   Estimates:
    Coefficients: [1x3 table]
      Covariance: [3x3 table]

ranovatbl = ranova(rm)

Thanks, Maria

[Todor]

Hi Maria,

try using this: rm = fitrm(t,'s1-s3~1')

Sincerely, Todor