How do I exclude anomalies from a plot in a data set?

6 次查看（过去 30 天）
JJH
JJH 2019-1-4
编辑： madhan ravi 2019-1-4
I want to plot a data set and exclude anomalous points. Say I have some data, e.g.
x = [0.5 0.48 0.2 0.51 0.49 0.52]
y = [1 2 3 4 5 6]
and I want to do a linear fit to the data excluding the anomalous point (x = 0.2). I can currently get matlab to find anomalous points as follows:
plot(x,y,'.','color', CM(ij,:));
fitA = polyfit(x,y,1);
fdata = polyval(fitA,x);
I = abs(fdata - y) > 1.2*std(y);
outliers = excludedata(x,y,'indices',I);
but I then want to do another fit where I exclude the outlying points. I tried
fitB = polyfit(x,y,1,'Exclude',outliers);
but this gave the error:
Error using polyfit
Too many input arguments.
How can I do this second fit correctly?

请先登录，再回答此问题。

采纳的回答

madhan ravi
madhan ravi 2019-1-4
编辑：madhan ravi 2019-1-4
https://www.mathworks.com/help/matlab/ref/polyfit.html - polyfit takes only three inputs
perhaps:
fitB(~ismember(fitB,outliers))

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Get Started with Curve Fitting Toolbox 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by