S09ABF (PDF version)
S Chapter Contents
S Chapter Introduction
NAG Library Manual

NAG Library Routine Document


Note:  before using this routine, please read the Users' Note for your implementation to check the interpretation of bold italicised terms and other implementation-dependent details.

+ Contents

    1  Purpose
    7  Accuracy

1  Purpose

S09ABF returns the value of the inverse circular cosine, arccosx, via the function name; the result is in the principal range 0,π.

2  Specification

REAL (KIND=nag_wp) S09ABF
REAL (KIND=nag_wp)  X

3  Description

S09ABF calculates an approximate value for the inverse circular cosine, arccosx. It is based on the Chebyshev expansion
where -12x 12,   and  t=4x2-1.
For x2 12,  arccosx= π2-arcsinx.
For -1x< -12,  arccosx=π-arcsin1-x2.
For 12<x1,  arccosx=arcsin1-x2.
For x>1,  arccosx is undefined and the routine fails.

4  References

Abramowitz M and Stegun I A (1972) Handbook of Mathematical Functions (3rd Edition) Dover Publications

5  Parameters

1:     X – REAL (KIND=nag_wp)Input
On entry: the argument x of the function.
Constraint: X1.0.
2:     IFAIL – INTEGERInput/Output
On entry: IFAIL must be set to 0, -1​ or ​1. If you are unfamiliar with this parameter you should refer to Section 3.3 in the Essential Introduction for details.
For environments where it might be inappropriate to halt program execution when an error is detected, the value -1​ or ​1 is recommended. If the output of error messages is undesirable, then the value 1 is recommended. Otherwise, if you are not familiar with this parameter, the recommended value is 0. When the value -1​ or ​1 is used it is essential to test the value of IFAIL on exit.
On exit: IFAIL=0 unless the routine detects an error or a warning has been flagged (see Section 6).

6  Error Indicators and Warnings

If on entry IFAIL=0 or -1, explanatory error messages are output on the current error message unit (as defined by X04AAF).
Errors or warnings detected by the routine:
S09ABF has been called with X>1.0, for which arccos is undefined. A zero result is returned.

7  Accuracy

If δ and ε are the relative errors in the argument and the result, respectively, then in principle
ε x arccosx 1-x2 ×δ .
The equality should hold if δ is greater than the machine precision (δ is due to data errors etc.), but if δ is due simply to round-off in the machine it is possible that rounding etc. in internal calculations may lose one extra figure.
The behaviour of the amplification factor xarccosx1-x2  is shown in the graph below.
In the region of x=0 this factor tends to zero and the accuracy will be limited by the machine precision. For x close to one, 1-xδ, the above analysis is not applicable owing to the fact that both the argument and the result are bounded x1, 0arccosxπ.
In the region of x-1 we have εδ, that is the result will have approximately half as many correct significant figures as the argument.
In the region x+1, we have that the absolute error in the result, E, is given by Eδ, that is the result will have approximately half as many decimal places correct as there are correct figures in the argument.
Figure 1
Figure 1

8  Further Comments


9  Example

This example reads values of the argument x from a file, evaluates the function at each value of x and prints the results.

9.1  Program Text

Program Text (s09abfe.f90)

9.2  Program Data

Program Data (s09abfe.d)

9.3  Program Results

Program Results (s09abfe.r)

S09ABF (PDF version)
S Chapter Contents
S Chapter Introduction
NAG Library Manual

© The Numerical Algorithms Group Ltd, Oxford, UK. 2012