How to Interpolated data outliers

7 views (last 30 days)
I have a data column that has 3630 rows of data. I used the method: [A, B] = rmoutliers (C, 'movmedian', window); to remove the data outliers.
These data were removed and I need them to be interpolated by the average of the entire column of data. How can I do this?

Accepted Answer

Turlough Hughes
Turlough Hughes on 2 Mar 2020
Edited: Turlough Hughes on 2 Mar 2020
Generate some sample data:
C = (1:3630).';
idxout = randperm(3630,500); % random index for outliers
C(idxout) = C(idxout)+2000*(1-2*rand(numel(idxout),1));
Using rmoutliers some of the values are removed:
window = 50;
[A, B] = rmoutliers(C, 'movmedian', window);
Seeing as you've indicated that you only have a column vector of data I assume you want to interpolate using index position as your input for x, you could try the following:
idx = find(~B); % C(idx) is the same as A
D = interp1(idx,A,1:size(C,1)); % Interpolated using index position as x values.
Let me know.
  1 Comment
Turlough Hughes
Turlough Hughes on 3 Mar 2020
Did this work for you? If not I suggest uploadibg the variable C as a .mat

Sign in to comment.

More Answers (1)

Matheus Brito
Matheus Brito on 4 Mar 2020
Hi Turlough
Yes, it worked very well and I was able to perform the procedures I needed.
Thank you!
  3 Comments
Matheus Brito
Matheus Brito on 9 Mar 2020
Hi Turlough Hughes, how are you?
Can you help me again? After interpolation, in some cases matlab gave me the following information:
Error using eig
Input matrix contains NaN or Inf.
Do you know why this may be happening? And how can I solve this? So to calculate some variables such as area, amplitude, frequency and others?
Thanks

Sign in to comment.

Categories

Find more on Interpolation in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!