NaN with backslash operator

Question

Sun 2019-4-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/459245-nan-with-backslash-operator

编辑： Sun 2019-7-31

采纳的回答： Adam Danz

Hi,

I would like to understand the reason as to why the below operation is giving a NaN solution for the attached data.

x = A\B;

Thank you.

YK..

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Adam Danz 2019-4-29

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/459245-nan-with-backslash-operator#answer_372871

在 MATLAB Online 中打开

The following rows of your variable "B" contain NaN values.

 find(isnan(B))
ans =
        1491
        1492
        1493
        1494
        1495
        1496
        1497

To ignore the NaN values,

nanIdx = isnan(A) | isnan(B); 
A(~nanIdx) \ B(~nanIdx)
ans =
    -0.091457

2 个评论
显示无隐藏无

John D'Errico 2019-4-29

The problem being that NaNs propagate. Everything they touch, turns to NaN. (Like old King MIdas, but in an even worse way.) So if you already have NaNs in your data, give it up. You need to excise them from your data before any linear algebra computations.

Steven Lord 2019-4-29

FYI the functions listed in the "Missing Data and Outliers" section of this documentation page may be of use in locating, removing, and/or replacing the NaN values in the data.

请先登录，再进行评论。