0
$\begingroup$

These two theorems are quite similar, they both show that an IVP has a unique point. Is there any difference between them.

To show an IVP has a solution, I usually show that the map is a contraction, thus IVP has a unique solution by Banach contraction principle. But if we apply global Picard theorem, there is no need to proving T is a contraction (i.e. as long as T is Lipschitz, it's fine.

$\endgroup$
0
$\begingroup$

The immediate use of the Banach fixed point theorem only works up to a finite time. Specifically, you define a map $A$ and seek its fixed point as the solution, and you find something like $\| A(f) - A(g) \| \leq LT \| f - g \|$ where $L$ is the Lipschitz constant from the equation and $T$ is the length of the interval you are using to define $A$. This will only be a contraction for $T<1/L$, so this argument only gives existence/uniqueness up to a finite time.

Global Picard amounts to performing the Banach fixed point theorem argument repeatedly to get $T$ to be larger than $1/L$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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