1Test of anorm(x) vs double series expansion


   2000 Random arguments were tested from the interval ( -0.663,  0.663)


  ANORM(X) was larger   140 times,
               agreed  1727 times, and
          was smaller   133 times.


 There are  24 base   2 significant digits in a floating-point number


 The maximum relative error of     0.2282E-06 =    2 ** -22.06
    occurred for X =-0.639754E+00
 The estimated loss of base   2 significant digits is   1.94


 The root mean square relative error was     0.3608E-07 =    2 ** -24.72
 The estimated loss of base   2 significant digits is   0.00





 Test of anorm(x) vs Taylor series about x-1/2


   2000 Random arguments were tested from the interval ( -5.657, -0.663)


  ANORM(X) was larger   978 times,
               agreed   389 times, and
          was smaller   633 times.


 There are  24 base   2 significant digits in a floating-point number


 The maximum relative error of     0.6656E-06 =    2 ** -20.52
    occurred for X =-0.474847E+01
 The estimated loss of base   2 significant digits is   3.48


 The root mean square relative error was     0.1773E-06 =    2 ** -22.43
 The estimated loss of base   2 significant digits is   1.57





 Test of anorm(x) vs Taylor series about x-1/2


   2000 Random arguments were tested from the interval (-12.438, -5.657)


  ANORM(X) was larger  1021 times,
               agreed   473 times, and
          was smaller   506 times.


 There are  24 base   2 significant digits in a floating-point number


 The maximum relative error of     0.4773E-06 =    2 ** -21.00
    occurred for X =-0.578060E+01
 The estimated loss of base   2 significant digits is   3.00


 The root mean square relative error was     0.1451E-06 =    2 ** -22.72
 The estimated loss of base   2 significant digits is   1.28


1Special Tests


       Check of identity anorm(X) + anorm(-X) = 1.0

         X            ANORM(-x)   ANORM(x)+ANORM(-x)-1

    0.766659E+01    0.883148E-14    0.000000E+00

    0.237098E+01    0.887044E-02    0.000000E+00

    0.910353E+01    0.437194E-19    0.000000E+00

    0.310169E+01    0.962093E-03    0.000000E+00

    0.122276E+00    0.451340E+00    0.000000E+00

    0.509703E+01    0.172509E-06    0.000000E+00

    0.583719E+01    0.265448E-08    0.000000E+00

    0.559129E+01    0.112695E-07    0.000000E+00

    0.279029E+01    0.263308E-02    0.000000E+00

    0.632259E+01    0.128604E-09    0.000000E+00



 Test of special arguments


 ANORM ( 0.340282E+39) =  0.100000E+01


 ANORM ( 0.000000E+00) =  0.500000E+00


 ANORM (-0.340282E+39) =  0.000000E+00


 Test of Error Returns



 ANORM will be called with the argument -0.971246E+01
 The result should not underflow


 ANORM returned the value 0.133410E-21



 ANORM will be called with the argument -0.129500E+02
 The result may underflow


 ANORM returned the value 0.000000E+00



 This concludes the tests
