How can I remove outliers in my data using Cook's Distance?

4 次查看（过去 30 天）

Fatemah Ebrahim 2020-6-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/556672-how-can-i-remove-outliers-in-my-data-using-cook-s-distance

编辑： Fatemah Ebrahim 2020-6-29

I have a large dataset, 6 .'xlsx' files with ~ 400,000 rows each, and I want to use Cook's Distance to determine the outliers in the fourth column of each dataset and then delete the corresponding row. How would I do that?

2 个评论
显示无隐藏无

KSSV 2020-6-29

https://in.mathworks.com/help/stats/cooks-distance.html

Fatemah Ebrahim 2020-6-29

编辑：Fatemah Ebrahim 2020-6-29

在 MATLAB Online 中打开

Hi! So I'm using the code they used on one of the '.xlsx' files as so:

X = A_t; % where this is a datetime value
Y = Adata(:,4); % where we are pulling the fourth column of the table
mdl = fitlm(X,Y);
plotDiagnostics(mdl,'cookd')
find((mdl.Diagnostics.CooksDistance)>3*mean(mdl.Diagnostics.CooksDistance))

And I am getting this error:

Error using classreg.regr.TermsRegression/handleDataArgs (line 550)

Predictor variables must be numeric vectors, numeric matrices, or

categorical vectors.

Error in LinearModel.fit (line 1184)

[X,y,haveDataset,otherArgs] =

LinearModel.handleDataArgs(X,varargin{:});

Error in fitlm (line 121)

model = LinearModel.fit(X,varargin{:});

Please let me know if you have any idea how to address this error, there does not seem to be much information on this. Thanks!

请先登录，再进行评论。

请先登录，再回答此问题。