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.

Question:

Let $ f$ be a strictly increasing, continuous function mapping $ I=[0,1]$ onto itself. Prove that the following inequality holds for all pairs $ x,y \in I$: $$ 1-\cos (xy) \le\int_0^xf(t) \sin {(tf(t))}dt + \int_0^y f^{-1}(t) \sin{(tf^{-1}(t))} dt .$$

For this problem I can want use Young inequality http://2000clicks.com/mathhelp/IneqYoungsInequality.aspx

so let $$g(t)=f(t)\sin{(tf(t))},g(0)=0$$

But $g(t)$ is strictly increasing function ? and $g^{-1}(t)=?$

so I can't use Young inequality to slove this problem.Maybe can use other methods? Thank you

This problem is from:http://www.artofproblemsolving.com/Forum/viewtopic.php?p=1390433&sid=fd9e67731e5084a02adb9974cf035c51#p1390433

share|improve this question

2 Answers 2

Denote \begin{align} T_1 &:=\{(u,v)\,|\,0\leq u\leq x,\,0\leq v\leq y\},\\ T_2 &:=\{(u,v)\,|\,0\leq u\leq x,\,0\leq v\leq f(u)\},\\ T_3 &:=\{(u,v)\,|\,0\leq v\leq y,\,0\leq u < f^{-1}(v)\}. \end{align} Then $T_1\subset T_2\cup T_3$, $T_2\cap T_3=\emptyset$ and all the three sets are Borel sets. So for every $g:[0,1]^2\to\mathbf{R}$, $g\geq 0$, Lebesgue-integrable function we have $$ \int_0^y\int_0^x g(u,v)\,du\,dv \leq \int_0^x\int_0^{f(u)} g(u,v)\,dv,du+\int_0^y\int_0^{f^{-1}(v)} g(u,v)\,du\,dv. $$ Taking $$ g(x,y):=\frac{\partial^2(1-\cos(xy))}{\partial y\partial x}=\sin(xy)+xy\cos(xy)\geq 0 $$ we obtain $$ 1-\cos (xy) \le\int_0^xf(u) \sin {(u\cdot f(u))}du + \int_0^y f^{-1}(v) \sin{(v\cdot f^{-1}(v))} dv. $$

share|improve this answer

From a geometric perspective, we need to prove that the right hand side is larger than than the area of the rectangle $xy$, and hench larger than $1-\cos(xy)$.If the integrands were simply $f(t)$ and $f'(t)$, the diagram below would suffice. Since $\sin$ is monotonous as well as $t f(t)$, you should be able to formulate a similar argument even though the integrands are no longer true inverses of each other. $$$$ $\quad\quad\quad\quad\quad\quad\quad\quad$enter image description here

share|improve this answer

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.