FLOAT scalar function The FLOAT function returns a floating-point representation of a number. FLOAT is a synonym for DOUBLE. FLOAT(numeric-expression ) The schema is SYSIBM.