Finding which rows in table contain NaN, save those as new table

Im looking to find a way to figure our which rows contain a 'NaN' in column 5, and then save all of those that do as a new table

[David Ebert]

Good question, same thing I'm trying to do.

Answer 1

[Guillaume]

Assuming your input is a table:

nantable = yourtable(isnan(yourtable{:, 5}), :)  %keep all rows whose 5th column is nan

Answer 2

[Adam Danz]

If the tables all contain the same variable names (headers), use the variable names to index columns rather than column number. This is one of the strengths of tables over matrices and cell arrays. If the variable names are different, then you'll have to use numeric indexing.

Suppose column 5 is named "MyData" and the table is named "T".

nanIdx = isnan(T.MyData); 
% If you must use numeric indexing, 
% nanIdx = isnan(T{:,5}); 

Create the new table

Tnew = T(nanIdx,:);

[David Ebert]

Yes!

[David Ebert]

Thanks!