Curve fitting tool - extract modified data

2 次查看（过去 30 天）

显示更早的评论

Anna Go 2019-1-24

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/441333-curve-fitting-tool-extract-modified-data

编辑： Anna Go 2019-1-24

在 MATLAB Online 中打开

Hi everyone,

I'm using the curve fitting tool for linear regression and to remove outliers for my data.

Below is an example of what I'm trying to do.

How can I recieve the new data for y where the one outier identified has been removed?

Thanks

xdat = [1; 2; 3; 4; 5; 6; 7; 8; 9; 10]
ydat = [ 100; 85; 66; 50; 40; 70; 25; 20; 15; 2]
%scatter(xdat,ydat)
f = fittype( 'poly1' );
% Fit model to data.
fit1 = fit( xdat, ydat, f);
% Link with code: https://de.mathworks.com/help/curvefit/removing-outliers.html
fdata = feval(fit1,xdat); 
%Set standard deviation to 0.5 
% Identifies outliers with deviation from baseline model
I = abs(fdata - ydat) > 0.5*std(ydat); 
outliers = excludedata(xdat,ydat,'indices',I);
% refit model to data with outliers excluded
fit2 = fit(xdat,ydat,f, 'Exclude',outliers)