Ignore NaN and -9999 values

Hi,

I have a table of data. There are 5 columns. 2000 rows.

Some data has NaN, 0 and -9999 values.

I want to calculate the average of each column, ignoring NaN, 0 and -9999 values.

What code can do this?

Thank you.

Answer 1

[BN]

Hi, It should be something like this, I learned it yesterday.

I assume T is the name of your table:

T_new = standardizeMissing(T, 0);
T_new = standardizeMissing(T_new, -9999)
A = T_new(:,vartype('numeric'));
omean = @(x) mean(x,'omitnan'); %ignoring NaN
mean_values = varfun(omean,A)

[Steven Lord]

You don't need to call standardizeMissing twice. It accepts a vector of values that should be converted to the standard missing value (which is NaN for double precision arrays.) Using a slightly modified version of the first example from its documentation page:

A = [0 1 5 -99 8 3 4 -99 16];
B = standardizeMissing(A,[-99, 0]);
beforeAndAfter = [A; B]

The 0 and the two -99 values in A were replaced by NaN in B.

[BN]

Thank you for let me know.

Answer 2

[Ameer Hamza]

% generting an example table
t = [1 2 3 4 5;
     0 2 1 3 -9999;
     -9999 5 7 0 9;
     nan 1 2 3 -9999];
T = array2table(t);

values = T.Variables;
mask = isnan(values) | values==0 | values==-9999;
values(mask) = nan;
result = nanmean(values, 1)

[BN]

+1