According to my lecture notes:

The properties of stability or attraction do not depend on the chosen initial time $t_0$. This is a consequence of the theorem of continuous dependence with respect to the initial conditions applied to the interval $[t_1,t_0]$ if $t_1$ is another initial time.

to give a bit of context stability and the notion of region of attraction is defined in my case for autonomous equations $x' = f(x)$. This is the theorem they're making reference to:

Let $D \subseteq \mathbb{R} \times \mathbb{R}^d$ open, $f:D \to \mathbb{R}^d$ continuous such that:

$\forall (t_0,x_0) \in D.$ (P) has a unique global solution.

Then, $\Omega = \{(t;t_0,x_0) \in \mathbb{R} \times \mathbb{R} \times \mathbb{R}^d. (t_0,x_0) \in D, \alpha(t_0,x_0) < t < \omega(t_0,x_0)\}$ is open and $G$ is continuous.

Do you see how to justify the above statement?


After reviewing the question I realize that it is easier to use the uniform convergence of solutions. Namely:

If $(t_{0n},x_{0n},\lambda_{0n}) \to (t_0,x_0,\lambda_0)$ and $\varphi_n$ are associated solutions with parameters $(t_{0n},x_{0n},\lambda_{0n})$ then for some large $n$ if we fix $I$ a compact subset of the domain of $\varphi$ the solution for parameters $(t_0,x_0,\lambda_0)$ then there is uniform convergence from $\varphi_n$ to $\varphi$ on I.

So something like this should work for stability:

enter image description here

Here $\varphi$ is the stable solution, the orange region represents the neighbourhood where solutions are guaranteed to stay withing the stability bounds, the pink curves represent the uniform convergence. I still need to formalize this reasoning.


Your Answer

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

Browse other questions tagged or ask your own question.