Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

$f$ and $g$ are two absolutely continous functions and $f$ is monotone.

Is $f(g)$ necessarily an absolutely continuous function? Why not any counter example?

Thanks a lot!!

share|improve this question
    
What did you try? –  Davide Giraudo Sep 13 '12 at 8:54
    
Refer to the book Measure Theory by V.I. Bogachev. There is hint for your problem. Try online books. :) –  B. S. Sep 13 '12 at 9:25
    
Babak I am curious about f(g) not g(f) thanks for the book though.. Davide I get stuck in the second stage when composed because disjointness criterion gets lost... –  Salih Ucan Sep 14 '12 at 1:29
add comment

2 Answers 2

up vote 2 down vote accepted

Let $\{x_n\}_1^\infty$, $\{y_n\}_1^\infty$ be two sequences of point in $I = [0, 1]$ such that $$ x_1 = 0\\ x_n\to 1\\ x_n < y_n < x_{n+1} \quad n=1, 2, \dotsc $$ Let's define a function $h$ on the interval I setting $$ h(x) = \begin{cases} \frac 1 {(y_n - x_n)(n + 1)^2} &x\in [x_n, y_n)\\ -\frac 1 {(x_{n + 1} - y_n)(n + 1)^2} &x\in [y_n, x_{n + 1}) \end{cases} $$ $h$ is summable $$ \begin{align} \int_I \lvert h(x)\rvert dx &= \sum_{n = 1}^\infty \frac {y_n - x_n} {(y_n - x_n)(n + 1)^2} + \sum_{n = 1}^\infty \frac {x_{n + 1} - y_n} {(x_{n + 1} - y_n)(n + 1)^2} \\ &= 2\sum_{n = 1}^\infty \frac 1 {(n + 1)^2} \\ &= \pi^2/3 - 2 \end{align} $$ Of course $$ f(x) := \int_0^x h(t) dt, \quad x\in I $$ is absolutely continuous, moreover $$ f(x_k) = \int_0^{x_k} h(t) dt = \sum_{n = 1}^{k - 1} \frac 1 {(n + 1)^2} - \sum_{n = 1}^{k - 1} \frac 1 {(n + 1)^2} = 0\\ f(y_k) = \int_0^{y_k} h(t) dt = \sum_{n = 1}^{k} \frac 1 {(n + 1)^2} - \sum_{n = 1}^{k - 1} \frac 1 {(n + 1)^2} = \frac 1 {(n + 1)^2} $$ From the above equalities we deduce the the total variation $V$ of the function $\sqrt f$ isn't finite $$ V \geq \sum_{n = 1}^{\infty} \left\vert \sqrt {f(y_n)} - \sqrt{f(x_n)} \right\vert = \sum_{n = 1}^{\infty} \frac 1 {n + 1} = \infty $$ In particular $\sqrt f$ is not absolutely continuous even if it is the composition of two absolutely continuous functions, $f$ and $$ g: x\in I \to \sqrt x \in I $$

share|improve this answer
add comment

Same idea as @AlbertH, but a bit simpler to write: $f(x)=x^2 \sin^2 \frac{\pi }{2x}$ is Lipschitz on $[0,1]$ because $f'$ is bounded. Hence, $f$ is absolutely continuous. However, $\sqrt{f}$ has infinite variation on $[0,1]$, which one can demonstrate by summing $|f(\frac1n)-f(\frac1{n+1})|$.

share|improve this answer
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.