Let $f(x,y) = 3(y-1)^{2/3y}$. I want to show that $y' = f(x,y)$ has two solutions for $y_0 = y(0) = 1$ and one solution for $y_0 = y(0) = 2$.

I am trying to understand the solution to this problem.

I understand all the computations that are done but I won't put all the details. just the idea of the solution

Let me start out by writing the theorem for the existence and uniqueness of the solution of the Cauchy Problem:

Let $U \in \mathbb{R}$ be an open subset. Let $f:U \rightarrow \mathbb{R^m}$ be a continuous function. $a,b \in \mathbb{R}_{>0}$ s.t. $A := \bar{B}_{\mathbb{R}}(x_0, a) \times \bar{B}_{\mathbb{R^n}}(y_0, b) \subset U.$ And $f\restriction_A$ is a lipschitz function. Then the Cauchy problem $y' = f(x,y), x(x_0) = y_0$ has one solution $\psi: [x_0-\alpha, x_0 + \alpha] \rightarrow \mathbb{R^m}$ where $\alpha = min(a, b/M), M= max_{x,y \in A} ||f(x,y)||_{\mathbb{R^m}}$.

  • So to start the solution, they want to check that $f$ is a lipschitz function in the neighborhood of the the interval $(0,2)$

I suppose, A in the theorem is $(0,2)$ here, in which case, how was the interval chosen

  • In order to . do so, let us consider the function $f$ on $[-1,1] \times [\dfrac{3}{2},\dfrac{5}{2}]$

Where do the intervals $[-1,1]$ and $[\dfrac{3}{2},\dfrac{5}{2}]$ come from?

  • Now, in order to show that it's a lipschitz function, I apply the following theorem: Let $f: [a,b] \rightarrow \mathbb{R}$ continuous. $\exists \xi \in [a,b] $ s.t. $f(b)-f(a) = f'(\xi)(b-a)$ and $ |f(b) - f(a)| \leq max_{[a,b]} f'(\xi)(b-a)$

  • We apply this theorem and we get that $|f(x,y)-f(x,z)| \leq max_{[3/2,5/2]}\dfrac{1}{(w-1)^{3/2}}|y-z|$

The commputations I understand.

  • Now let us show that $f$ is not a lipschitz function on $(0,1)$

Once again where does $(0,1)$ come from and why do we have to show that f is not a lipschitz function on this interval$

  • Then the last step is simply to show that $y' = f(x,y)$ has two solutions for $y_0 = y(0) = 1$ and one solution for $y_0 = y(0) = 2$.

Could someone try to explain and maybe claify the different steps of the solution


Your main confusion is to think that $(0,1)$ and $(0,2)$, given in your explanation, are intervals. In fact, they are the points of the initial condition of the Cauchy Problem, i.e. $(x_0,y_0) = (0,1)$ and $(x_0,y_0) = (0,2)$ for both questions in your statement.

For the second case, the one that you want to prove the uniqueness of the Cauchy Problem $$ \begin{cases} f(x,y) = 3(y-1)^{2/3y} \\ y_0 = y(0) = 2 \end{cases}, $$ you must find the set $A$ of the theorem to use it. This set is $A = [-1,1] \times \left[\dfrac{3}{2},\dfrac{5}{2}\right]$. Notice that, using the notation of your enunciation of the theorem, \begin{align*} &\bar{B}_{\mathbb{R}}(x_0, a) = \bar{B}_{\mathbb{R}}(0, 1) = [-1,1]\\ &\bar{B}_{\mathbb{R^n}}(y_0, b) = \bar{B}_{\mathbb{R}}(2, 1/2) = \left[\dfrac{3}{2},\dfrac{5}{2}\right], \end{align*} where here your dimension $n$ of the second ball is $n=1$ (you only have one dependent variable, $y$), $a$ and $b$ are the radious of the balls (in your case simply intervals) $1$ and $1/2$, respectively.


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.