Could non-smooth time-limited functions been Analytical? Please read the scenarios first
I was reading about analytic functions definitions on Wiki and looks like some of its properties where Differentiabilty is involved depends on where is the function defined on $\mathbb{R}$ or $\mathbb{C}$, and also reading about Bump functions $\in C_c^\infty$ on Wiki and Non-analytic smooth functions here, it is said that "One can easily prove that any analytic function of a real argument is smooth. The converse is not true" and also if a function is smooth and compact-supported, then it cannot be an analytic function.
So here is where I get kind of lost: if I think in $f(t)$ as a continuous time-limited function (so is compact-supported with boundaries $\partial t = \{t_0,\,t_F\}$, and also, so is continuous and compact-supported, it is also a bounded function)... since is compacted supported, if I let it restricted to be a "smooth function" I will already know it couldn't be analytical, ¿but what will happen if is not smooth?
So, to understand what could happen, given differentiability could be an domain-related issue, I want to know what could happen in each of the following scenarios:
- Let $f(t)$ been a continuous time-limited function such as $f(t)\in\mathbb{R}$ that is not smooth (even, it could being neither a differentiable function), fulfilling also that $f(t_0)=f(t_F)=0$... Could this $f(t)$ been an analytical function?
- Let $f(t)$ been a continuous time-limited function such as $f(t)\in\mathbb{R}$ that is not smooth (even, it could being neither a differentiable function), where it can have any of their values at the edges of the domain $f(t_0)\neq 0$ and/or $f(t_F)\neq 0$ (not necessarily both equals, so $f(t_0) \neq f(t_F)$ in general)... Could this $f(t)$ been an analytical function?
- Let $f(t)$ been a continuous time-limited function such as $f(t)\in\mathbb{C}$ that is not smooth (even, it could being neither a differentiable function), fulfilling also that $f(t_0)=f(t_F)=0$... Could this $f(t)$ been an analytical function?
- Let $f(t)$ been a continuous time-limited function such as $f(t)\in\mathbb{C}$ that is not smooth (even, it could being neither a differentiable function), where it can have any of their values at the edges of the domain $f(t_0)\neq 0$ and/or $f(t_F)\neq 0$ (not necessarily both equals, so $f(t_0) \neq f(t_F)$ in general)... Could this $f(t)$ been an analytical function?
PS: the "edges-part" is because if they are non-zero, it cannot be smooth and compacted-supported under no-other conditions, and also since they made a discontinuity in the "transition", it creates some issues in its derivatives (where some delta functions arise), and also in their Fourier transforms, and I am interested if this issue has something to said within the topic of this question.
As example, $$f(t) = \begin{cases} \cos^4\left(\frac{t\pi}{2}\right),\,|t|\leq 1 \\ 0,\,|t|>1 \end{cases}$$ is a real valued function with compact-support that is not smooth (because is not true that $\lim\limits_{t \to \partial t} \frac{df^n(t)}{dt^n} = 0\,\forall n \geq 0$), but it is analytical "within" its support (i.e., within the open interval $(-1,\,1)$), and its transition to the boundaries is "soft" (meaning here, that the $f(t)$ and $f'(t)$ are continuous and equal to zero at $t = \{-1,\,1\}$ with $\lim\limits_{t \to \partial t^-} f(t) = \lim\limits_{t \to \partial t^+} f(t)$ and $\lim\limits_{t \to \partial t^-} f'(t) = \lim\limits_{t \to \partial t^+} f'(t)$), here the plot of $f(t)$, it looks actually really "good behaved", so I want to know if these kind of functions could also be analytic - or if not.