The following function $f(t,x):[0,T]\times R\mapsto R$ such that $\int^T_0|f(t,0)|^2 d t<\infty$, where $0<T<\infty$.

If $f(t,x)$ satisfies $|f(t,x)|\leq Ax+B$ for each $x\in R$ and $A, B$ are given constants, then we say that $f(t,x)$ is of linear growth in $x$. All Lipschitz functions are of linear growth.

If $f(t,x)$ satisfies $|f(t,x)|\leq C+C|x|^p$ for each $x\in R$, $C>0$ and $p\geq1$ are given constants, then we say $f(t,x)$ is of polynominal growth in $x$.

Besides, if there exists a strictly increasing convex function $\varphi(\cdot)$ with $\varphi(0)=\varphi(0+)$ such that for each $x\in R$, $|f(t,x)|\leq g_t+\varphi(x)$, where $g_t$ is suqare integrable with respect to $t$, then we say $f(t,x)$ is convex growth in $x$. Convex growth functions contain many types of functions, such as exponential functions, tangent functions, even linear growth functions and polynominal growth functions.

Yet there exists a general growth. We say $f(t,x)$ has a general growth if for each $\alpha\geq 0$, $\sup_{|x|\leq \alpha}|f(t,x)-f(t,0)|\in L^1([0,T])$, where $L^1$ means Lebesgue integrable.

I am very confused that from theoretical point, the general growth is actually more general than the convex growth, but I can not find an explicit function satisfies the general growth other than the convex growth. All functions I can find can be controlled by a convex function, i.e., they are all convex growth.

Anyone can help me to find a function with general growth but not convex growth?

Here is my example: $e^{-xt^2}$. I do not know if it is right.

  • $\begingroup$ Something's not right. For each fixed $\alpha$, $\sup_{|x|\le \alpha} |f(x) - f(0)|$ is a number, not a function of anything, so what is the variable with respect to which it is supposed to be $L^1$? Do you mean there is a locally $L^1$ function $g(\alpha)$ such that $\sup_{|x|\le \alpha} |f(x) - f(0)| \le g(\alpha)$? It can't be globally $L^1$ if you allow any growth at all. $\endgroup$ – Robert Israel Aug 29 '15 at 1:42
  • $\begingroup$ @RobertIsrael Sorry, my mistake. Another variable, time t, should be considered. $\endgroup$ – XIAO Lishun Aug 29 '15 at 2:08
  • $\begingroup$ Your example is bounded by $1$. $\endgroup$ – Robert Israel Aug 30 '15 at 4:08

You might try $f(t,x) = t^{-1/2} x^2$ for $t > 0$, $f(0,x) = 0$.


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.