Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Reading a book on fractional calculus reminded me that I'd like to know more on the following method/idea.

Given an integral: $\displaystyle \int_{-\pi}^\pi{\frac{\displaystyle\sum_{j=a}^b{e^{i\cdot c_j \cdot t}}}{\displaystyle\sum_{k=a}^b{e^{i \cdot d_k \cdot t}}} dt}$

I can use recurrences and generating functions to find all of the derivatives. Using all of the derivatives (evaluated at a point), I can essentially create another function (for example, the function at point $n$ is the $n$th derivative). The function brings up a curiousity: how accurate could it be at predicting the integral itself?

EDIT An additional thought: If this method fails, perhaps the derivatives could still be used to aid in a very good spline extrapolation using only a few points.

So I'm wondering what problems arise when trying to extrapolate/interpolate an integral of the form above using derivatives. I'd like to know any studies of this problem if they exist, and what I can expect. I realize I'm being a bit vague, but I'm not exactly sure what the good questions are regarding this issue. I would like to explore this possibility as much as possible, and hope that answers arrive with hope and how to evaluate the integral this way.

share|cite|improve this question
A ratio of sums of complex exponentials? That's going to be rather oscillatory on the real line, depending on the values of the $c_j$ and the $d_k$ . Maybe constructing a nice contour (avoiding any poles that might occur, of course) might help here? – J. M. Nov 10 '10 at 15:28
up vote 1 down vote accepted

If you can calculate all the derivatives at a point, say t=0, you can create a Laurent series around zero. As long as all the poles of the function are outside a circle of radius $\pi$, this will converge to the function. You can then integrate term by term to get the integral.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.