How do i remove an outlier?

Question

Wouter 2023-1-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1892200-how-do-i-remove-an-outlier

编辑： Wouter 2023-1-11

采纳的回答： Constantino Carlos Reyes-Aldasoro

I have a data set where there is an outlier in the date. I am trying to remove it but the way i am trying is not working. The way i am trying it is by writing that if the next number is so much more higher than the previous it should change it to the previous one. I might not have understood correctly how this works because it is not working. Any help is appreciated!

clear all

close all

clc

load('ACE2-opdracht-2-persoonlijk.mat', 'meetdata');

A = meetdata;

Motortoerental = A (:,1);

Stuurspanning = A (:,2);

Tijd = (0:4000)./50;

figure;

subplot(1,2,1)

plot(Tijd, Motortoerental); hold on

xlabel('Tijd(s)')

ylabel('Toerental(omw/min)')

title('v1');

grid;

subplot(1,2,2)

plot(Tijd, Stuurspanning); hold on

xlabel('Tijd(s)')

ylabel('Stuurspanning(V)')

title('v2');

grid;

Stoorsignaal = 0.02*sin(2*pi*Tijd);

v2_verbeterd = Stuurspanning - Stoorsignaal;

figure;

plot(Tijd,v2_verbeterd);

xlabel('Tijd(s)');

ylabel('Stuurspanning(V)');

grid;

% This is the part where i try to remove the outlier. I include my whole code just to be sure though.

for i = 2:length(Motortoerental)

if ((Motortoerental(i)-Motortoerental(i-1))/0.1)>5

Motortoerental_verbeterd(i) = y(i-1);

end

figure;

plot(Tijd,Motortoerental_verbeterd);

xlabel('Tijd(s)')

ylablel('Toerental(omw/min)')

grid;

3 个评论
显示 1更早的评论隐藏 1更早的评论

Wouter 2023-1-11

I must have done something wrong. But when i enter that line of code it doesn't remove the outlier from my plot. Do you have any further suggestions?

Dyuman Joshi 2023-1-11

在 MATLAB Online 中打开

What is y here? It has not been defined in the code above

for i = 2:length(Motortoerental)
    if ((Motortoerental(i)-Motortoerental(i-1))/0.1)>5
        Motortoerental_verbeterd(i) = y(i-1);
    end
end

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Constantino Carlos Reyes-Aldasoro 2023-1-11

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1892200-how-do-i-remove-an-outlier#answer_1146535

在 MATLAB Online 中打开

Use a moving filter, in your case, the best would be to replace each value by the median value of a neighbourhood, that is if you have [1 3 2] the median is 2, if you have [1 1000 2] median is also 2, check this code

a=sin(0:0.1:100);

a(55)=3;

plot(a)

b=movmedian(a,3);

plot(b)

2 个评论
显示无隐藏无

Dyuman Joshi 2023-1-11

在 MATLAB Online 中打开

In doing so, the data gets skewed. You can see that the values have changed

a=sin(0:0.1:100);
c=a;
a(55)=3;
b=movmedian(a,3);
b==c
ans = 1×1001 logical array
   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   0   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   1   0   1   1   1   1   1   1   1   1   1   0
nnz(b~=c)
ans = 36