2025-02-14 | | Total: 5
We establish a non-explosion result for rough differential equations (RDEs) in which both the noise and drift coefficients together with their derivatives are allowed to grow at infinity. Additionally, we prove the existence of a bi-continuous solution flow for stochastic differential equations (SDEs). In the case of RDEs with additive noise, we show that our result is optimal by providing a counterexample.
We propose to relax the classic Cauchy-Riemann equations for a mapping. We support the interest of such a proposal by looking at one specific situation in 3D, and proving the existence of pairs of harmonic conjugate functions with respect to a unitary gradient as the title of this contribution conveys. We further investigate the relationship between boundary conditions for such pairs, the importance of the unitary constraint, and the eventual link of these ideas to Calderón's problem in 3D.
Let $f=\sum_{n=0}^\infty f_n x^n \in \overline{\mathbb Q}[[x]$ be a solution of an algebraic differential equation $Q(x,y(x), \ldots, y^{(k)}(x))=0$, where $Q$ is a multivariate polynomial with coefficients in $\overline{\mathbb Q}$. The sequence $(f_n)_{n\ge 0}$ satisfies a non-linear recurrence, whose expression involves a polynomial $M$ of degree $s$. When the equation is linear, $M$ is its indicial polynomial at the origin. We show that when $M$ is split over $\mathbb Q$, there exist two positive integers $\delta$ and $\nu$ such that the denominator of $f_n$ divides $\delta^{n+1}(\nu n+\nu)!^{2s}$ for all $n\ge 0\ $, generalizing a well-known property when the equation is linear. This proves in this case a strong form of a conjecture of Mahler that Pólya--Popken's upper bound $n^{\mathcal{O}(n\log(n))}$ for the denominator of $f_n$ is not optimal. This also enables us to make Sibuya and Sperber's bound $\vert f_n\vert_v\le e^{\mathcal{O}(n)}$, for all finite places $v$ of $\overline{\mathbb Q}$, explicit in this case. Our method is completely effective and rests upon a detailed $p$-adic analysis of the above mentioned non-linear recurrences. Finally, we present various examples of differentially algebraic functions for which the associated polynomial $M$ is split over $\mathbb Q$, among which are Weierstra\ss' elliptic $\wp$ function, solutions of Painlevé equations, and Lagrange's solution to Kepler's equation.
Let $\Omega \subset \mathbb{R}$ be a nonempty and open set, then for all $f, g, h\in \mathscr{C}^{2}(\Omega)$ we have \begin{multline*} \frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}}(f\cdot g\cdot h) -f\frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}}(g\cdot h)-g\frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}}(f\cdot h)-h\frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}}(f\cdot g) + f\cdot g\frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}} h+f\cdot h\frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}} g+g\cdot h \frac{\mathrm{d}^{2}}{\mathrm{d} x^{2}}f=0 \end{multline*} The aim of this paper is to consider the corresponding operator equation \[ D(f\cdot g \cdot h) - fD(g\cdot h) - gD(f\cdot h) - hD(f \cdot g) + f\cdot g D(h) + f\cdot h D(g) +g\cdot h D(f) =0 \] for operators $D\colon \mathscr{C}^{k}(\Omega)\to \mathscr{C}(\Omega)$, where $k$ is a given nonnegative integer and the above identity is supposed to hold for all $f, g, h \in \mathscr{C}^{k}(\Omega)$. We show that besides the operators of first and second derivative, there are more solutions to this equation. Some special cases characterizing differential operators are also studied.
In this article we consider a modification of the Stein's spherical maximal operator of complex order $\alpha$ on ${\mathbb R^n}$: $$ {\mathfrak M}^\alpha_{[1,2]} f(x) =\sup\limits_{t\in [1,2]} \big| {1\over \Gamma(\alpha) } \int_{|y|\leq 1} \left(1-|y|^2 \right)^{\alpha -1} f(x-ty) dy\big|. $$ We show that when $n\geq 2$, suppose $\|{\mathfrak M}^{\alpha}_{[1,2]} f \|_{L^q({\mathbb R^n})} \leq C\|f \|_{L^p({\mathbb R^n})}$ holds for some $\alpha\in \mathbb{C}$, $p,q\geq1$, then we must have that $q\geq p$ and $${\rm Re}\,\alpha\geq \sigma_n(p,q):=\max\left\{\frac{1}{p}-\frac{n}{q},\ \frac{n+1}{2p}-\frac{n-1}{2}\left(\frac{1}{q}+1\right),\frac{n}{p}-n+1\right\}.$$ Conversely, we show that ${\mathfrak M}^\alpha_{[1,2]}$ is bounded from $L^p({\mathbb R^n})$ to $L^q({\mathbb R^n})$ provided that $q\geq p$ and ${\rm Re}\,\alpha>\sigma_2(p,q)$ for $n=2$; and ${\rm Re}\,\alpha>\max\left\{\sigma_n(p,q), 1/(2p)- (n-2)/(2q) -(n-1)/4\right\}$ for $n>2$. The range of $\alpha,p$ and $q$ is almost optimal in the case either $n=2$, or $\alpha=0$, or $(p,q)$ lies in some regions for $n>2$.