nearfloat

Arguments

dir

string ("succ" or "pred")

x

real scalar, vector or matrix

xnear

real scalar, vector or matrix

Description

This function computes, in the element wise meaning, the corresponding neighbours of the elements of x (in the underlying floating point set, see number_properties), the successors if dir = "succ" and the predecessors if dir = "pred".

Examples

format("e",22)
nearfloat("succ",1) - 1
1 - nearfloat("pred",1)
format("v") //reset default format

See also

• number_properties — determine floating-point parameters
• frexp — dissect floating-point numbers into base 2 exponent and mantissa