The data type FLOAT (p) defines a floating point number (floating_point_literal). A column is defined that has a floating point number with precision p (0<p<=38).
The data type FLOAT is specified when columns are defined.