Computational Complexity of One-Step Methods for a Scalar Autonomous Differential Equation.
Abstract
The problem is to calculate an approximate solution of an initial value problem for a scalar autonomous differential equation. A generalized notion of a nonlinear Runge-Kutta (NRK) method is defined. We show that the order of any s-stage NRK method cannot exceed 2s-1; hence, the family of NRK methods due to Brent has the maximal order possible. Using this result, we derive complexity bounds on the problem of finding an approximate solution with error not exceeding epsilon. We also compute the order which minimizes these bounds, and show that this optimal order increases as epsilon decreases, tending to infinity as epsilon tends to zero. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1976
- Accession Number
- ADA035930
Entities
People
- Arthur G. Werschulz
Organizations
- Carnegie Mellon University