feq, fge, fgt, fle, flt, fne#

Purpose#

Fuzzy comparison functions. These functions use _fcmptol to fuzz the comparison operations to allow for roundoff error.

Format#

ret = feq(a, b)#

ret = fge(a, b)#

ret = fgt(a, b)#

ret = fle(a, b)#

ret = flt(a, b)#

ret = fne(a, b)#

Parameters:

a (NxK matrix) – first matrix.
b (LxM matrix) – second matrix, ExE compatible with a.

Returns:

ret (scalar) – returns 1 if TRUE and 0 if FALSE.

Global Input#

_fcmptol#: scalar, comparison tolerance. The default value is 1.0e-15.

Examples#

Example 1: Fuzzy equality#

_fcmptol = 1e-12;

ret = feq(2, 2 + 1e-13);

The above code will set ret equal to 1, because 2 and (2 + 1e-13) differ by less than the value of _fcmptol, which is 1e-12.

Example 2: Fuzzy greater than#

_fcmptol = 1e-10;

a = 0.5;
b = a + 1e-5;
c = a + 1e-11;

ret_1 = fgt(b, a);

ret_2 = fgt(c, a);

After the code above:

ret_1 = 1
ret_2 = 0

Remarks#

The return value is TRUE if every comparison is TRUE.

The statement:

ret = feq(a, b);

is equivalent to:

ret = abs(a-b) <= _fcmptol;

For the sake of efficiency, these functions are not written to handle missing values. If a and b contain missing values, use missrv() to convert the missing values to something appropriate before calling a fuzzy comparison function.

The calling program can reset _fcmptol before calling these procedures:

_fcmptol = 1e-12;

Source#

fcompare.src