Missing data in repeated measure model

I am using the matlab function fitrm to fit repeated measures model in order to investigate whether elements grouped according to Grouping1 have statistically different means for the variable measured at time t1,t2,t3,t4 (var_t1,var_t2,var_t3,var_t4).

My data look like the ones in the table: (better formatted here http://stackoverflow.com/questions/32475023/missing-data-in-repeated-measure-model)

      Grouping1   Grouping2     Gender     Age       BMI       var_t1           var_t2          var_t3        var_t4
      ______    ___________    ______    ______    ______    ____________    ____________    ____________    ____________

      C           B             Male       60       24.802    836             608             746             NaN         
      C           A             Male       67       19.818    242             544             460             483         
      ...       ...
      D           C             Female     65       21.631    621             468             NaN             NaN

As you can see from I have some missing data for var_t3 and var_t4. Can I still use fitrm?

If fit a repeated measures model, where var_t1,-var_t4 are the responses and Grouping1, Grouping2, Gender, Age and BMI are the predictor variables

    Time = [1:4]';    
    rm = fitrm(table,'var_t1-var_t4 ~ Grouping1 + Grouping2 + Gender + Age + BMI','WithinDesign',Time)

the function doesn't return error, but I don't know if the results have any meaning...

Answer 1

[Aditya]

Hi Gabrielle,

When using fitrm in MATLAB to fit a repeated measures model, handling missing data is an important consideration. MATLAB's fitrm function can handle missing data to some extent, but it's crucial to understand how it affects your results and what assumptions are being made.

Handling Missing Data:

1. Listwise Deletion: By default, fitrm uses listwise deletion for rows with any missing data in the response variables. This means that if any of the response variables (var_t1 to var_t4) have missing values for a particular observation, that entire observation will be excluded from the analysis.
2. Impact on Results: If missing data is not missing completely at random (MCAR), listwise deletion can introduce bias. It's important to consider the mechanism of missingness and whether the missing data might skew your results.

Example code:

% Define the within-subject factor
Time = table([1; 2; 3; 4], 'VariableNames', {'Time'});

% Fit the repeated measures model
rm = fitrm(dataTable, 'var_t1-var_t4 ~ Grouping1 + Grouping2 + Gender + Age + BMI', 'WithinDesign', Time);

% Display the results
ranovatbl = ranova(rm);
disp(ranovatbl);