欠損値をmasked arrayで表現すると，欠損値が含まれるデータの重みつき平均を取ることができる．たとえば，SSTの全球平均をとるのに便利．
numpy.ma.average(a, axis=None, weights=None, returned=False)
You can take a weighted average of the data that contains the missing values, using numpy masked array. This is useful, for example, to take a global average of SSTs.
numpy.ma.average (a, axis = None, weights = None, returned = False)