Numerical stability
= Numerical stability
{wiki=Numerical_stability}
Numerical stability refers to the behavior of algorithms in the presence of finite precision arithmetic, which is common in computer calculations. Specifically, it addresses how errors (such as rounding errors) can affect the results of numerical computations. An algorithm is considered numerically stable if small changes in the input (whether due to rounding errors or perturbations) lead to small changes in the output. Conversely, an algorithm that amplifies errors significantly is considered numerically unstable.