Can we sum a series of values,even some of them are "NaN"?If i want to do it,how can i do?

I get a series answer of optimal problem in several times,and i want to sum of them and average them,however, some of them are "NaN",can i or how to write a code to ignore those "NaN" and sum the others which is not "NaN" ?

The version of matlab is 2015a

Answer 1

[Walter Roberson]

You did not indicate your MATLAB version.

For a couple of years now, you have been able to write sum(TheArray, 'omitnan') .

Before that, the easiest approach was to use the Statistics toolbox nansum()

mean() these days also has 'omitnan' flag, and in older versions the Stats tool was nanmean()

[yang-En Hsiao]

Thanks for information,i have edited the question,if there is a vector call A

A=[1 2 3 4 5 6 7 8 9 NaN]
ANSwess=nansum(A)

The window will show me ANSwess=45,but why when i write the code as

A=[1 2 3 4 5 6 7 8 9 NaN]
ANSwess=nansum(A(1:9)+A(end))

The window will show me ANSwess=45,same result as sum(A)?

[Walter Roberson]

That last line returns 0 for me. A(end) is Nan, and adding nan to each A(1:9) entry gives nan for each result, so you would be applying nansum to an all-nan vector, and the result of that is going to be 0.