The most obvious way to approximate the derivative of a function numerically is to use the definition of derivative and stick in a small value of the step size h.
f′ (x) ≈ ( f(x + h) − f(x) ) / h.
How small should h be? Since the exact value of the derivative is the limit as h goes to zero, the smaller h is the better. Except that’s not true in computer arithmetic. When h is too small, f(x + h) is so close to f(x) that the subtraction loses precision.
One way to see this is to think of the extreme case of h so small that f(x + h) equals f(x) to machine precision. Then the derivative is approximated by 0, regardless of what the actual derivative is.
As h gets smaller, the approximation error decreases, but the numerical error increases, and so there is some optimal value of h.
You can do significantly better by using a symmetric approximation for the derivative:
f′ (x) ≈ ( f(x + h) − f(x − h) ) / 2h.
For a given value of h, this formula has about the same numerical error but less approximation error. You still have a trade-off between approximation error and numerical error, but it’s a better trade-off.
If the function f that you want to differentiate is analytic, i.e. differentiable as a function of a complex variable, you can take the step h to be a complex number. When you do, the numerical difficulty completely goes away, and you can take h much smaller.
Suppose h is a small real number and you take
f′ (x) ≈ ( f(x + ih) − f(x − ih) ) / 2ih.
Now f(x + ih) and −f(x − ih) are approximately equal by the Schwarz reflection principle. And so rather than canceling out, the terms in the numerator add. We have
f(x + ih) − f(x − ih) ≈ 2 f(x + ih)
and so
f′ (x) ≈ f(x + ih) / ih.
Example
Let’s take the derivative of the gamma function Γ(x) at 1 using each of the three methods above. The exact value is −γ where γ is the Euler-Mascheroni constant. The following Python code shows the accuracy of each approach.
from scipy.special import gamma def diff1(f, x, h): return (f(x + h) - f(x))/h def diff2(f, x, h): return (f(x + h) - f(x - h))/(2*h) γ = 0.5772156649015328606065 x = 1 h = 1e-7 exact = -γ approx1 = diff1(gamma, x, h) approx2 = diff2(gamma, x, h) approx3 = diff2(gamma, x, h*1j) print(approx1 - exact) print(approx2 - exact) print(approx3 - exact)
This prints
9.95565755390615e-08 1.9161483510998778e-10 (9.103828801926284e-15-0j)
In this example the symmetric finite difference approximation is about 5 times more accurate than the asymmetric approximation, but the complex step approximation is 10,000,000 times more accurate.
It’s a little awkward that the complex step approximation returns a complex number, albeit one with zero complex part. We can eliminate this problem by using the approximation
f′ (x) ≈ Im f(x + ih) / h
which will return a real number. When we implement this in Python as
def diff3(f, x, h): return (f(x + h*1j) / h).imag
we see that it produces the same result as diff2
but without the zero imaginary part.
This is a cool technique. I have never seen it before. It does seem to me that there is a bit of an apples-to-oranges comparison because there might be many more arithmetic operations involved in calculating the function of a complex variable versus the function of a real variable. Perhaps a better comparison would be to a higher-order purely-real formula. On the other hand, if one is comparing lines of code and coding effort, this really is an apples-to-apples comparison.
Your exposition is hard for me to follow at f(x + ih) − f(x − ih) ≈ 2 f(x + ih)
If f(x) = x, for example, is it not the case that
f(x+ih)-f(x-ih) = (x + ih) – (x – ih) = -2 ih. ?
and, of course, -2 ih / (-2 ih) = 1 = the derivative of the function f(x) = x
More generally, isn’t f(x + ih) − f(x − ih) = 2 ih (df/dx) + higher order terms?