Prove that $({aover a+b})^3+({bover b+c})^3+ ({cover c+a})^3geq {3over 8}$
$begingroup$
Let $a,b,c$ be positive real numbers. Prove that $$Big({aover a+b}Big)^3+Big({bover b+c}Big)^3+ Big({cover c+a}Big)^3geq {3over 8}$$
If we put $x=b/a$, $y= c/b$ and $z=a/c$ we get $xyz=1$ and
$$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq {3over 8}$$
Since $f(x)=Big({1over 1+x}Big)^3$ is convex we get, by Jensen,: $$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq 3f({x+y+zover 3})$$
Unfortunately, since $f$ is decreasing we don't have $f({x+y+zover 3}) geq f(1) = {1over 8}$.
Some idea how to solve this?
inequality cauchy-schwarz-inequality holder-inequality jensen-inequality uvw
$endgroup$
add a comment |
$begingroup$
Let $a,b,c$ be positive real numbers. Prove that $$Big({aover a+b}Big)^3+Big({bover b+c}Big)^3+ Big({cover c+a}Big)^3geq {3over 8}$$
If we put $x=b/a$, $y= c/b$ and $z=a/c$ we get $xyz=1$ and
$$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq {3over 8}$$
Since $f(x)=Big({1over 1+x}Big)^3$ is convex we get, by Jensen,: $$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq 3f({x+y+zover 3})$$
Unfortunately, since $f$ is decreasing we don't have $f({x+y+zover 3}) geq f(1) = {1over 8}$.
Some idea how to solve this?
inequality cauchy-schwarz-inequality holder-inequality jensen-inequality uvw
$endgroup$
4
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26
add a comment |
$begingroup$
Let $a,b,c$ be positive real numbers. Prove that $$Big({aover a+b}Big)^3+Big({bover b+c}Big)^3+ Big({cover c+a}Big)^3geq {3over 8}$$
If we put $x=b/a$, $y= c/b$ and $z=a/c$ we get $xyz=1$ and
$$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq {3over 8}$$
Since $f(x)=Big({1over 1+x}Big)^3$ is convex we get, by Jensen,: $$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq 3f({x+y+zover 3})$$
Unfortunately, since $f$ is decreasing we don't have $f({x+y+zover 3}) geq f(1) = {1over 8}$.
Some idea how to solve this?
inequality cauchy-schwarz-inequality holder-inequality jensen-inequality uvw
$endgroup$
Let $a,b,c$ be positive real numbers. Prove that $$Big({aover a+b}Big)^3+Big({bover b+c}Big)^3+ Big({cover c+a}Big)^3geq {3over 8}$$
If we put $x=b/a$, $y= c/b$ and $z=a/c$ we get $xyz=1$ and
$$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq {3over 8}$$
Since $f(x)=Big({1over 1+x}Big)^3$ is convex we get, by Jensen,: $$Big({1over 1+x}Big)^3+Big({1over 1+y}Big)^3+ Big({1over 1+z}Big)^3geq 3f({x+y+zover 3})$$
Unfortunately, since $f$ is decreasing we don't have $f({x+y+zover 3}) geq f(1) = {1over 8}$.
Some idea how to solve this?
inequality cauchy-schwarz-inequality holder-inequality jensen-inequality uvw
inequality cauchy-schwarz-inequality holder-inequality jensen-inequality uvw
edited Dec 11 '18 at 19:51
Xander Henderson
14.3k103554
14.3k103554
asked Dec 4 '18 at 16:51
greedoidgreedoid
40.7k1149100
40.7k1149100
4
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26
add a comment |
4
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26
4
4
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26
add a comment |
3 Answers
3
active
oldest
votes
$begingroup$
Let $g(t):=left(dfrac{1}{1+exp(t)}right)^3$ for $tinmathbb{R}$. Then, $$g''(t)=frac{3,exp(t),big(3,exp(t)-1big)}{big(1+exp(t)big)^5}text{ for each }tinmathbb{R},.$$
Thus, $g$ is convex on $big[-ln(3),inftybig)$.
Let $x:=dfrac{b}{a}$, $y:=dfrac{c}{b}$, and $z:=dfrac{a}{c}$ be as the OP defines. Then, the required inequality is equivalent to
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq dfrac{3}{8},.tag{*}$$
If $x$, $y$, or $z$ is less than $dfrac{1}{3}$, then clearly the left-hand side of (*) is greater than
$$left(dfrac{1}{1+frac13}right)^3=frac{27}{64}>frac38,.$$
If all $x$, $y$, and $z$ are greater than or equal to $dfrac13$, then $ln(x),ln(y),ln(z)geq -ln(3)$, so that we can use convexity of $g$ on $big[-ln(3),inftybig)$. By Jensen's Inequality,
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq3,gleft(frac{ln(x)+ln(y)+ln(z)}{3}right)=3,g(0)=frac{3}{8},.$$
Hence, the equality holds if and only if $x=y=z=1$, making $a=b=c$.
$endgroup$
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
add a comment |
$begingroup$
By Holder $$left(sum_{cyc}frac{a^3}{(a+b)^3}right)^2sum_{cyc}1geqleft(sum_{cyc}sqrt[3]{left(frac{a^3}{(a+b)^3}right)^2cdot1}right)^3=left(sum_{cyc}frac{a^2}{(a+b)^2}right)^3.$$
Thus, it's enough to prove that
$$frac{left(sumlimits_{cyc}frac{a^2}{(a+b)^2}right)^3}{3}geqfrac{9}{64}$$ or
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}geqfrac{3}{4}.$$
Now, by C-S
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}=sumlimits_{cyc}frac{a^2(a+c)^2}{(a+b)^2(a+c)^2}geqfrac{left(sumlimits_{cyc}(a^2+ab)right)^2}{sumlimits_{cyc}(a+b)^2(a+c)^2}.$$
Thus, it's enough to prove that
$$4left(sumlimits_{cyc}(a^2+ab)right)^2geq3sumlimits_{cyc}(a+b)^2(a+c)^2,$$
which is true even for all reals $a$, $b$ and $c$.
Indeed, the last inequality is symmetric inequality by degree four,
which says that by $uvw$ (https://math.stackexchange.com/tags/uvw/info )
it's enough to prove the last inequality for equality case of two variables and since
it's the homogeneous inequality by even degree, we can assume $b=c=1$, which gives
$$(a-1)^2(a+3)^2geq0.$$
Done!
$endgroup$
add a comment |
$begingroup$
This is more of a comment, but I don't have the reputation. Use Lagrange multipliers. Solving, you find that the critical points occur when $xyz=1$ and $yz (1+x)^4 = xz (1+y)^4 = xy (1+z)^4$. I think the only solution is $x=y=z=1$. Clearly, it's a minimum and plugging in shows the bound.
Added:
We can rewrite the condition as $xyz=1$ and $frac{(1+x)^4}{x} = frac{(1+y)^4}{y} = frac{(1+z)^4}{z}$
The function $g(x)=frac{(1+x)^4}{x}$ is decreasing from $0$ to $1/3$ and increasing from $1/3$ to $infty$. This shows that 2 of $x,y,z$ must be equal (WLOG $x$ and $y$) and $z=1/x^2$. It remains to solve $frac{(1+x)^4}{x} = frac{(1+1/x^2)^4}{1/x^2}$. This time, it's not hard to check $x=1$ is the only solution and we are done.
$endgroup$
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3025819%2fprove-that-a-over-ab3b-over-bc3-c-over-ca3-geq-3-over-8%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
3 Answers
3
active
oldest
votes
3 Answers
3
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
Let $g(t):=left(dfrac{1}{1+exp(t)}right)^3$ for $tinmathbb{R}$. Then, $$g''(t)=frac{3,exp(t),big(3,exp(t)-1big)}{big(1+exp(t)big)^5}text{ for each }tinmathbb{R},.$$
Thus, $g$ is convex on $big[-ln(3),inftybig)$.
Let $x:=dfrac{b}{a}$, $y:=dfrac{c}{b}$, and $z:=dfrac{a}{c}$ be as the OP defines. Then, the required inequality is equivalent to
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq dfrac{3}{8},.tag{*}$$
If $x$, $y$, or $z$ is less than $dfrac{1}{3}$, then clearly the left-hand side of (*) is greater than
$$left(dfrac{1}{1+frac13}right)^3=frac{27}{64}>frac38,.$$
If all $x$, $y$, and $z$ are greater than or equal to $dfrac13$, then $ln(x),ln(y),ln(z)geq -ln(3)$, so that we can use convexity of $g$ on $big[-ln(3),inftybig)$. By Jensen's Inequality,
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq3,gleft(frac{ln(x)+ln(y)+ln(z)}{3}right)=3,g(0)=frac{3}{8},.$$
Hence, the equality holds if and only if $x=y=z=1$, making $a=b=c$.
$endgroup$
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
add a comment |
$begingroup$
Let $g(t):=left(dfrac{1}{1+exp(t)}right)^3$ for $tinmathbb{R}$. Then, $$g''(t)=frac{3,exp(t),big(3,exp(t)-1big)}{big(1+exp(t)big)^5}text{ for each }tinmathbb{R},.$$
Thus, $g$ is convex on $big[-ln(3),inftybig)$.
Let $x:=dfrac{b}{a}$, $y:=dfrac{c}{b}$, and $z:=dfrac{a}{c}$ be as the OP defines. Then, the required inequality is equivalent to
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq dfrac{3}{8},.tag{*}$$
If $x$, $y$, or $z$ is less than $dfrac{1}{3}$, then clearly the left-hand side of (*) is greater than
$$left(dfrac{1}{1+frac13}right)^3=frac{27}{64}>frac38,.$$
If all $x$, $y$, and $z$ are greater than or equal to $dfrac13$, then $ln(x),ln(y),ln(z)geq -ln(3)$, so that we can use convexity of $g$ on $big[-ln(3),inftybig)$. By Jensen's Inequality,
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq3,gleft(frac{ln(x)+ln(y)+ln(z)}{3}right)=3,g(0)=frac{3}{8},.$$
Hence, the equality holds if and only if $x=y=z=1$, making $a=b=c$.
$endgroup$
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
add a comment |
$begingroup$
Let $g(t):=left(dfrac{1}{1+exp(t)}right)^3$ for $tinmathbb{R}$. Then, $$g''(t)=frac{3,exp(t),big(3,exp(t)-1big)}{big(1+exp(t)big)^5}text{ for each }tinmathbb{R},.$$
Thus, $g$ is convex on $big[-ln(3),inftybig)$.
Let $x:=dfrac{b}{a}$, $y:=dfrac{c}{b}$, and $z:=dfrac{a}{c}$ be as the OP defines. Then, the required inequality is equivalent to
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq dfrac{3}{8},.tag{*}$$
If $x$, $y$, or $z$ is less than $dfrac{1}{3}$, then clearly the left-hand side of (*) is greater than
$$left(dfrac{1}{1+frac13}right)^3=frac{27}{64}>frac38,.$$
If all $x$, $y$, and $z$ are greater than or equal to $dfrac13$, then $ln(x),ln(y),ln(z)geq -ln(3)$, so that we can use convexity of $g$ on $big[-ln(3),inftybig)$. By Jensen's Inequality,
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq3,gleft(frac{ln(x)+ln(y)+ln(z)}{3}right)=3,g(0)=frac{3}{8},.$$
Hence, the equality holds if and only if $x=y=z=1$, making $a=b=c$.
$endgroup$
Let $g(t):=left(dfrac{1}{1+exp(t)}right)^3$ for $tinmathbb{R}$. Then, $$g''(t)=frac{3,exp(t),big(3,exp(t)-1big)}{big(1+exp(t)big)^5}text{ for each }tinmathbb{R},.$$
Thus, $g$ is convex on $big[-ln(3),inftybig)$.
Let $x:=dfrac{b}{a}$, $y:=dfrac{c}{b}$, and $z:=dfrac{a}{c}$ be as the OP defines. Then, the required inequality is equivalent to
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq dfrac{3}{8},.tag{*}$$
If $x$, $y$, or $z$ is less than $dfrac{1}{3}$, then clearly the left-hand side of (*) is greater than
$$left(dfrac{1}{1+frac13}right)^3=frac{27}{64}>frac38,.$$
If all $x$, $y$, and $z$ are greater than or equal to $dfrac13$, then $ln(x),ln(y),ln(z)geq -ln(3)$, so that we can use convexity of $g$ on $big[-ln(3),inftybig)$. By Jensen's Inequality,
$$gbig(ln(x)big)+gbig(ln(y)big)+gbig(ln(z)big)geq3,gleft(frac{ln(x)+ln(y)+ln(z)}{3}right)=3,g(0)=frac{3}{8},.$$
Hence, the equality holds if and only if $x=y=z=1$, making $a=b=c$.
edited Dec 4 '18 at 21:46
answered Dec 4 '18 at 21:42
BatominovskiBatominovski
1
1
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
add a comment |
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I don't get it. What is the difference with my solution?
$endgroup$
– greedoid
Dec 4 '18 at 21:45
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
$begingroup$
I'm not sure what to say to that, but my idea was to bypass the inequality $frac{x+y+z}{3}geq sqrt[3]{xyz}$ that you would need to use in your attempt.
$endgroup$
– Batominovski
Dec 4 '18 at 21:48
add a comment |
$begingroup$
By Holder $$left(sum_{cyc}frac{a^3}{(a+b)^3}right)^2sum_{cyc}1geqleft(sum_{cyc}sqrt[3]{left(frac{a^3}{(a+b)^3}right)^2cdot1}right)^3=left(sum_{cyc}frac{a^2}{(a+b)^2}right)^3.$$
Thus, it's enough to prove that
$$frac{left(sumlimits_{cyc}frac{a^2}{(a+b)^2}right)^3}{3}geqfrac{9}{64}$$ or
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}geqfrac{3}{4}.$$
Now, by C-S
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}=sumlimits_{cyc}frac{a^2(a+c)^2}{(a+b)^2(a+c)^2}geqfrac{left(sumlimits_{cyc}(a^2+ab)right)^2}{sumlimits_{cyc}(a+b)^2(a+c)^2}.$$
Thus, it's enough to prove that
$$4left(sumlimits_{cyc}(a^2+ab)right)^2geq3sumlimits_{cyc}(a+b)^2(a+c)^2,$$
which is true even for all reals $a$, $b$ and $c$.
Indeed, the last inequality is symmetric inequality by degree four,
which says that by $uvw$ (https://math.stackexchange.com/tags/uvw/info )
it's enough to prove the last inequality for equality case of two variables and since
it's the homogeneous inequality by even degree, we can assume $b=c=1$, which gives
$$(a-1)^2(a+3)^2geq0.$$
Done!
$endgroup$
add a comment |
$begingroup$
By Holder $$left(sum_{cyc}frac{a^3}{(a+b)^3}right)^2sum_{cyc}1geqleft(sum_{cyc}sqrt[3]{left(frac{a^3}{(a+b)^3}right)^2cdot1}right)^3=left(sum_{cyc}frac{a^2}{(a+b)^2}right)^3.$$
Thus, it's enough to prove that
$$frac{left(sumlimits_{cyc}frac{a^2}{(a+b)^2}right)^3}{3}geqfrac{9}{64}$$ or
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}geqfrac{3}{4}.$$
Now, by C-S
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}=sumlimits_{cyc}frac{a^2(a+c)^2}{(a+b)^2(a+c)^2}geqfrac{left(sumlimits_{cyc}(a^2+ab)right)^2}{sumlimits_{cyc}(a+b)^2(a+c)^2}.$$
Thus, it's enough to prove that
$$4left(sumlimits_{cyc}(a^2+ab)right)^2geq3sumlimits_{cyc}(a+b)^2(a+c)^2,$$
which is true even for all reals $a$, $b$ and $c$.
Indeed, the last inequality is symmetric inequality by degree four,
which says that by $uvw$ (https://math.stackexchange.com/tags/uvw/info )
it's enough to prove the last inequality for equality case of two variables and since
it's the homogeneous inequality by even degree, we can assume $b=c=1$, which gives
$$(a-1)^2(a+3)^2geq0.$$
Done!
$endgroup$
add a comment |
$begingroup$
By Holder $$left(sum_{cyc}frac{a^3}{(a+b)^3}right)^2sum_{cyc}1geqleft(sum_{cyc}sqrt[3]{left(frac{a^3}{(a+b)^3}right)^2cdot1}right)^3=left(sum_{cyc}frac{a^2}{(a+b)^2}right)^3.$$
Thus, it's enough to prove that
$$frac{left(sumlimits_{cyc}frac{a^2}{(a+b)^2}right)^3}{3}geqfrac{9}{64}$$ or
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}geqfrac{3}{4}.$$
Now, by C-S
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}=sumlimits_{cyc}frac{a^2(a+c)^2}{(a+b)^2(a+c)^2}geqfrac{left(sumlimits_{cyc}(a^2+ab)right)^2}{sumlimits_{cyc}(a+b)^2(a+c)^2}.$$
Thus, it's enough to prove that
$$4left(sumlimits_{cyc}(a^2+ab)right)^2geq3sumlimits_{cyc}(a+b)^2(a+c)^2,$$
which is true even for all reals $a$, $b$ and $c$.
Indeed, the last inequality is symmetric inequality by degree four,
which says that by $uvw$ (https://math.stackexchange.com/tags/uvw/info )
it's enough to prove the last inequality for equality case of two variables and since
it's the homogeneous inequality by even degree, we can assume $b=c=1$, which gives
$$(a-1)^2(a+3)^2geq0.$$
Done!
$endgroup$
By Holder $$left(sum_{cyc}frac{a^3}{(a+b)^3}right)^2sum_{cyc}1geqleft(sum_{cyc}sqrt[3]{left(frac{a^3}{(a+b)^3}right)^2cdot1}right)^3=left(sum_{cyc}frac{a^2}{(a+b)^2}right)^3.$$
Thus, it's enough to prove that
$$frac{left(sumlimits_{cyc}frac{a^2}{(a+b)^2}right)^3}{3}geqfrac{9}{64}$$ or
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}geqfrac{3}{4}.$$
Now, by C-S
$$sumlimits_{cyc}frac{a^2}{(a+b)^2}=sumlimits_{cyc}frac{a^2(a+c)^2}{(a+b)^2(a+c)^2}geqfrac{left(sumlimits_{cyc}(a^2+ab)right)^2}{sumlimits_{cyc}(a+b)^2(a+c)^2}.$$
Thus, it's enough to prove that
$$4left(sumlimits_{cyc}(a^2+ab)right)^2geq3sumlimits_{cyc}(a+b)^2(a+c)^2,$$
which is true even for all reals $a$, $b$ and $c$.
Indeed, the last inequality is symmetric inequality by degree four,
which says that by $uvw$ (https://math.stackexchange.com/tags/uvw/info )
it's enough to prove the last inequality for equality case of two variables and since
it's the homogeneous inequality by even degree, we can assume $b=c=1$, which gives
$$(a-1)^2(a+3)^2geq0.$$
Done!
answered Dec 4 '18 at 21:35
Michael RozenbergMichael Rozenberg
101k1591193
101k1591193
add a comment |
add a comment |
$begingroup$
This is more of a comment, but I don't have the reputation. Use Lagrange multipliers. Solving, you find that the critical points occur when $xyz=1$ and $yz (1+x)^4 = xz (1+y)^4 = xy (1+z)^4$. I think the only solution is $x=y=z=1$. Clearly, it's a minimum and plugging in shows the bound.
Added:
We can rewrite the condition as $xyz=1$ and $frac{(1+x)^4}{x} = frac{(1+y)^4}{y} = frac{(1+z)^4}{z}$
The function $g(x)=frac{(1+x)^4}{x}$ is decreasing from $0$ to $1/3$ and increasing from $1/3$ to $infty$. This shows that 2 of $x,y,z$ must be equal (WLOG $x$ and $y$) and $z=1/x^2$. It remains to solve $frac{(1+x)^4}{x} = frac{(1+1/x^2)^4}{1/x^2}$. This time, it's not hard to check $x=1$ is the only solution and we are done.
$endgroup$
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
add a comment |
$begingroup$
This is more of a comment, but I don't have the reputation. Use Lagrange multipliers. Solving, you find that the critical points occur when $xyz=1$ and $yz (1+x)^4 = xz (1+y)^4 = xy (1+z)^4$. I think the only solution is $x=y=z=1$. Clearly, it's a minimum and plugging in shows the bound.
Added:
We can rewrite the condition as $xyz=1$ and $frac{(1+x)^4}{x} = frac{(1+y)^4}{y} = frac{(1+z)^4}{z}$
The function $g(x)=frac{(1+x)^4}{x}$ is decreasing from $0$ to $1/3$ and increasing from $1/3$ to $infty$. This shows that 2 of $x,y,z$ must be equal (WLOG $x$ and $y$) and $z=1/x^2$. It remains to solve $frac{(1+x)^4}{x} = frac{(1+1/x^2)^4}{1/x^2}$. This time, it's not hard to check $x=1$ is the only solution and we are done.
$endgroup$
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
add a comment |
$begingroup$
This is more of a comment, but I don't have the reputation. Use Lagrange multipliers. Solving, you find that the critical points occur when $xyz=1$ and $yz (1+x)^4 = xz (1+y)^4 = xy (1+z)^4$. I think the only solution is $x=y=z=1$. Clearly, it's a minimum and plugging in shows the bound.
Added:
We can rewrite the condition as $xyz=1$ and $frac{(1+x)^4}{x} = frac{(1+y)^4}{y} = frac{(1+z)^4}{z}$
The function $g(x)=frac{(1+x)^4}{x}$ is decreasing from $0$ to $1/3$ and increasing from $1/3$ to $infty$. This shows that 2 of $x,y,z$ must be equal (WLOG $x$ and $y$) and $z=1/x^2$. It remains to solve $frac{(1+x)^4}{x} = frac{(1+1/x^2)^4}{1/x^2}$. This time, it's not hard to check $x=1$ is the only solution and we are done.
$endgroup$
This is more of a comment, but I don't have the reputation. Use Lagrange multipliers. Solving, you find that the critical points occur when $xyz=1$ and $yz (1+x)^4 = xz (1+y)^4 = xy (1+z)^4$. I think the only solution is $x=y=z=1$. Clearly, it's a minimum and plugging in shows the bound.
Added:
We can rewrite the condition as $xyz=1$ and $frac{(1+x)^4}{x} = frac{(1+y)^4}{y} = frac{(1+z)^4}{z}$
The function $g(x)=frac{(1+x)^4}{x}$ is decreasing from $0$ to $1/3$ and increasing from $1/3$ to $infty$. This shows that 2 of $x,y,z$ must be equal (WLOG $x$ and $y$) and $z=1/x^2$. It remains to solve $frac{(1+x)^4}{x} = frac{(1+1/x^2)^4}{1/x^2}$. This time, it's not hard to check $x=1$ is the only solution and we are done.
edited Dec 4 '18 at 21:44
answered Dec 4 '18 at 20:56
zoidbergzoidberg
1,065113
1,065113
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
add a comment |
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
2
2
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
$begingroup$
It seems to me there's enough here to post as an answer, no need to apologize.
$endgroup$
– David K
Dec 4 '18 at 21:11
add a comment |
Thanks for contributing an answer to Mathematics Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3025819%2fprove-that-a-over-ab3b-over-bc3-c-over-ca3-geq-3-over-8%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
4
$begingroup$
How is this question "seeking personal advice," whoever voted to close this thread?
$endgroup$
– Batominovski
Dec 4 '18 at 21:26