I would assume you want an output the same size as the input, in which case:
V = [7 8 95 210 85 NaN NaN 52 NaN]' %demo data
result = nan(size(v));
result(~isnan(v)) = factorial(v(~isnan(v)))
As documented factorial is only exact for input less than or equal to 21. factorial([95, 85, 52]) gives you an inaccurate result but with the correct order of magnitude. factorial(210) is far beyound what can be stored as double precision. Anything above factorial(171) is too big.
