How to Delete entries for which the mid-quote deviated by more than 10 mean absolute deviations from a rolling centered median (excluding the observation under consideration) of 50 observations (25 observations before and 25 after).

I have a cell array of midquotes split on days. I need to Delete entries for which the mid-quote deviated by more than 10 mean absolute deviations from a rolling centered median (excluding the observation under consideration) of 50 observations (25 observations before and 25 after) for EACH DAY. How to do it with R2016a.

[NS]

any help on this . i need to remove outlier from each day using this .

Answer 1

[KSSV]

Let A be your data vector, which has outliers and you want to remove:

idx =  abs(A - mean(T)) > 3*std(T);  % get indices 
A(idx) = [] ;   % remove them 

YOu can run a loop or use Cellfun to use the above lines of code to remove outliers in a cell.

[NS]

Sir I have a table A with multiple columns out of which i need to perform this on column named MIDQUOTE. In your above answer , what is T ?

[NS]

Any help ? Does the above codes includes rolling window of 50 ?

[NS]

I have a cell array A which is indexed to DATETIME index idx and i need to perform this function on 8th column of each cell of cell array . I ran this code and got Index exceeds matrix dimension error .

[idx1] =  cellfun(@(x){abs(x - mean(x) > 3*std(x))},A(:,8),'UniformOutput',false);

Kindly help how to remove the outliers from each cell within this code ?

[NS]

Kindly help ?