Numerical integration/Adaptive Simpson's method: Difference between revisions

From Rosetta Code
Content added Content deleted
m (Fix wp link)
m (added zkl header)
Line 32: Line 32:
'''return''' _quad_asr(f, a, fa, b, fb, tol, whole, m, fm, depth)
'''return''' _quad_asr(f, a, fa, b, fb, tol, whole, m, fm, depth)
|}
|}

=={{header|zkl}}==
<lang zkl></lang>
<lang zkl></lang>
{{out}}
<pre>
</pre>

Revision as of 22:36, 29 September 2018

Numerical integration/Adaptive Simpson's method is a draft programming task. It is not yet considered ready to be promoted as a complete task, for reasons that should be found in its talk page.

Lychee (1969)'s Modified Adaptive Simpson's method (doi:10.1145/321526.321537) is a numerical quadrature method that recursively bisects the interval until the precision is high enough.

Pseudocode: Simpson's method, adaptive
; Lychee's ASR, Modifications 1, 2, 3
procedure _quad_asr_simpsons(f, a, fa, b, fb)
    m := (a + b) / 2
    fm := f(m)
    h := b - a
    return multiple [m, fm, (h / 6) * (f(a) + f(b) + 4*sum1 + 2*sum2)]

procedure _quad_asr(f, a, fa, b, fb, tol, whole, m, fm, depth)
    lm, flm, left  := _quad_asr_simpsons(f, a, fa, m, fm)
    rm, frm, right := _quad_asr_simpsons(f, m, fm, b, fb)
    delta := left + right - whole
    
    tol' := tol / 2
    if depth <= 0 or tol' == tol or abs(delta) <= 15 * tol:
        return left + right + delta / 15
    else:
        return _quad_asr(f, a, fa, m, fm, tol', left , lm, flm, depth - 1) +
               _quad_asr(f, m, fm, b, fb, tol', right, rm, frm, depth - 1)

procedure quad_asr(f, a, b, tol, depth)
   fa := f(a)
   fb := f(b)
   m, fm, whole := _quad_asr_simpsons(f, a, fa, b, fb)
   return _quad_asr(f, a, fa, b, fb, tol, whole, m, fm, depth)

zkl

<lang zkl></lang> <lang zkl></lang>

Output: