Dropping NaN observations from Matrix
显示 更早的评论
Hey There.
So i have a matrix that has a person ID for the first column, then data in the rows for each. What i need to do is i have identified that there are some observations that are NaN, so i need to write a code that identifies which person ID have NaN observations, so that i can drop those observations from the data.
Any help could be appreciated
回答（2 个）
Walter Roberson
2019-1-23
0 个投票
With R2016b or later see https://www.mathworks.com/help/matlab/ref/rmmissing.html
Kevin Phung
2019-1-23
编辑：Kevin Phung
2019-1-23
if you had a matrix A and wanted to search for NaNs in each row
nanRows =[];
for i = 1:size(A,1)
if any(isnan(A(i,:))) % for each row, if any of the elements contain a NaN
nanRows(i) = i;% store that index
end
end
A(nanRows,:) = []; %remove those IDs from your matrix
if your matrix is of cell arrays, then replace parentheses with {}.
edit: before you run A(nanRows,:) =[] , you can check which IDs had Nans with just A(nanRows,1)
2 个评论
Walter Roberson
2019-1-23
No loop needed.
A(any(isnan(A),2),:) = [];
Kevin Phung
2019-1-23
never noticed any() can take two arguments, that's neat! Thanks Walter.
类别
在 帮助中心 和 File Exchange 中查找有关 NaNs 的更多信息
2019-1-23
2019-1-23
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
