In the mathematical subfield of numerical analysis, numerical stability is a desirable property of numerical algorithms. The precise definition of stability depends on the context, but it is derived from the accuracy of the algorithm.
An opposite phenomenon is instability. Typically, algorithms would approach the right solution in the limit, if there were no round-off or truncation errors, but depending on the specific computational method, errors can be magnified, instead of damped, causing the error to grow exponentially.
Sometimes a single calculation can be achieved in several ways, all of which are algebraically equivalent in terms of ideal real or complex numbers, but in practice when performed on digital computers yield different results. Some calculations might damp out approximation errors that occur; others might magnify such errors. Calculations that can be proven not to magnify approximation errors are called numerically stable. One of the common tasks of numerical analysis is to try to select algorithms which are robust – that is to say, have good numerical stability among other desirable properties.
Read more about Numerical Stability: Example, Forward, Backward, and Mixed Stability, Error Growth, Stability in Numerical Differential Equations
Famous quotes containing the words numerical and/or stability:
“The moment a mere numerical superiority by either states or voters in this country proceeds to ignore the needs and desires of the minority, and for their own selfish purpose or advancement, hamper or oppress that minority, or debar them in any way from equal privileges and equal rights—that moment will mark the failure of our constitutional system.”
—Franklin D. Roosevelt (1882–1945)
“The world can be at peace only if the world is stable, and there can be no stability where the will is in rebellion, where there is not tranquility of spirit and a sense of justice, of freedom, and of right.”
—Woodrow Wilson (1856–1924)