In decimal floating-point operations, a value, not interpreted as a mathematical value, that contains a mask and a sequence of floating-point digits and that causes an invalid operation condition if used in certain arithmetic operations. See also decimal floating-point number.