Can someone prove or disproof the claim:

Given a locally lipschitz vector field $f$ with associated ODE $\dot x = f(x)$, then the solution $x$ must be locally lipschitz

Note: local lipschitz condition:

$\exists r, L$ s.t. $\forall x,y, |x - x_o | < r, |y - y_o| < r \Rightarrow |f(x) - f(y)| \leq L |x-y|$

I tried some examples:

$\dot x = 1 \Rightarrow x = t$. Then $f$ is locally lipschitz (in fact globally), and $x$ is locally lipschitz

$\dot x = x \Rightarrow x = x_o \exp(t)$. $f$ is locally lipschitz (but not globally), and $x$ is locally lipschitz (since$ x \in C^1$)

$\dot x = x^2$ and higher ups are a bit too involved...

If the claim holds, does it also hold for global lipschitz vector fields?

  • $\begingroup$ Do you mean $x$ locally Lipschitz in $t$? $\endgroup$ – John B Feb 28 '16 at 22:30
  • $\begingroup$ The solution is continuously differentiable and hence locally L. $\endgroup$ – Artem Feb 28 '16 at 22:31

$x$ is differentiable, therefore continuous; $f$ is locally Lipschitz, therefore continuous. Therefore $\dot{x} = f(x)$ is continuous, i.e. $x$ is continuously differentiable. Continuously differentiable functions are locally Lipschitz.


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.