Matrix exponential

Matrix exponential algorithm with improved numerical precision
Updated 19 Mar 2014

View License

In the classical scale-and-square exponential algorithm, numeric precision is degraded by large diagonal terms in matrices that are nearly equal to the identity matrix. This variant algorithm separates out the identity matrix terms, preserving high precision with arbitrarily small scaling factors.
[Requires polyvalm_, File ID: #45786.]

Cite As

Kenneth Johnson (2024). Matrix exponential (, MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2013a
Compatible with any release
Platform Compatibility
Windows macOS Linux
Find more on Matrix Exponential in Help Center and MATLAB Answers
Tags Add Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
Version Published Release Notes

Updated test cases to use MATLAB R2014A. (Previous version used R2013A.)

Spelling error in comments.

Corrected output specification in comment header.

Replaced two corrupted comment lines.

Noted requirement for polyvalm_ in Description.

No change - just trying to fix a problem in File Exchange UI.

- Added minusI optional input.
- Revised p polynomial definition to include leading I term.
- Use polyvalm_ to do the polynomial evaluation.
- Use more robust method for setting m and s.
- Modified test cases in comment header.

Added acknowledgment of Federal support and disclaimer.

Added reference to another good test case in program comments.