Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Does there exist a concave, increasing function $h\colon[0,\infty)\to\mathbb{R}$ such that

  1. $\lim_{x\to\infty} h(x)=\infty$

  2. $\lim_{x\to\infty} h(x)/x=0$

  3. There exist sequences of positive numbers $a_n,b_n,c_n,d_n$ which converge to infinity such that:

3a. $\infty>\lim_{n\to\infty} a_n/b_n=\lim_{n\to\infty} c_n/d_n>0$ but

3b. $\lim_{n\to\infty} h(a_n)/h(b_n) \neq \lim_{n\to\infty} h(c_n)/h(d_n)$

Is it possible?

share|improve this question
    
Could you edit your text so that latex becomes readable? See the "How to write Math" text on this page. Thanks. –  András Bátkai Oct 19 '12 at 9:37
    
Thank You for suggestion –  user27381 Oct 19 '12 at 10:03
    
this is a type of stackexchange question, isn't it? –  John B Oct 19 '12 at 11:08
    
well I need it in the stacking-cuutting construction of rank one systems –  user27381 Oct 19 '12 at 12:30
add comment

1 Answer

The negation of your property 3. is called regular variation.

Here is an example of a concave not regularly varying function (taken from this paper by Iksanov and Rösler, p.10) :

Take $$f(x) = 2^{-k} x + 2^{k+1} - 3, \;\;\;\; x \in [4^k,4^{k+1})$$

Then for $x_n=4^n$, $y_n=3\cdot4^n$, $$\lim_n \frac{f(2x_n)}{f(x_n)} = 2 \neq \frac{7}{5} = \lim_n \frac{f(2y_n)}{f(y_n)}.$$

share|improve this answer
    
Thank You very much for suggestion and for to drawing my attention to this paper –  user27381 Oct 19 '12 at 12:33
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.