Prove for all positive reals a,b,c,d: $ \frac{a-b}{b+c}+\frac{b-c}{c+d}+\frac{c-d}{d+a}+\frac{d-a}{a+b} \geq 0$
Problem
Source: Middle Europe Mathematical olympiad TST Day 1 First problem
Tags: inequalities, LaTeX, inequalities unsolved
18.06.2009 15:52
Matematika wrote: Prove for all positive reals a,b,c,d: $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} \geq 0$ Let $ a\geq b \geq c \geq d$ then $ LHS \geq \frac{a-b}{a+b} +\frac{b-c}{a+b} +\frac{c-d}{a+b} +\frac{d-a}{a+b}=\frac{0}{a+b}=0$
18.06.2009 16:04
You aren't allowed to assume $ a \geq b\geq c\geq d$ cause inequality isn't symmetric. As much as I know you can't assume that if inequality is cyclic.
18.06.2009 16:45
Why not , do you have any mathematical explanation for it ?
18.06.2009 17:27
enndb0x wrote: Matematika wrote: Prove for all positive reals a,b,c,d: $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} \geq 0$ Let $ a\geq b \geq c \geq d$ then $ LHS \geq \frac {a - b}{a + b} + \frac {b - c}{a + b} + \frac {c - d}{a + b} + \frac {d - a}{a + b} = \frac {0}{a + b} = 0$ Can you repeat your argument when I assume $ a\ge c\ge b\ge d$?.My solution : The inequality is equivalent to \[ \frac{a+c}{b+c}+\frac{b+d}{c+d}+\frac{c+a}{d+a}+\frac{d+b}{a+b}\ge 4 \] Apply Cauchy Schwarz inequality in Engel form, we have \[ LHS\ge \frac{(2(a+b+c+d))^2}{(a+c)(b+c)+(b+d)(c+d)+(c+a)(d+a)+(d+b)(a+b)} \] It remains to prove \[ \frac{(2(a+b+c+d))^2}{(a+c)(b+c)+(b+d)(c+d)+(c+a)(d+a)+(d+b)(a+b)} \ge 4 \] But in the fact: we have \[ (a+b+c+d)^2=(a+c)(b+c)+(b+d)(c+d)+(c+a)(d+a)+(d+b)(a+b) \] It ends our proof.
18.06.2009 19:22
Can you elaborate more how you won that equivalence ?
19.06.2009 01:44
that was mine solution on competition as well. mathematical proof for that assuming is simply counter example you have expression 2(a-b) (it is cyclic for 2 arguments) and if you could assume $ a \geq b$ you would get that it is bigger then 0 and you can take b less than a and that's not true. and that assumption comes from the fact that in symmetric inequality if you change two variables you don't change a thing but in cyclic ones you do change inequality. just one comment you don't need to add 4 cauchy in engel form is correct for negative values of nominators (denominator must be positive)
27.06.2009 23:20
There is faster way to prove $ \frac {a + c}{b + c} + \frac {b + d}{c + d} + \frac {c + a}{d + a} + \frac {d + b}{a + b}\ge 4$. We have \[ \frac {a + c}{b + c} + \frac {b + d}{c + d} + \frac {c + a}{d + a} + \frac {d + b}{a + b} = (a + c)(\frac {1}{b + c} + \frac {1}{d + a}) + (b + d)(\frac {1}{c + d} + \frac {1}{a + b})\geq \frac {4(a + b + c + d)}{a + b + c + d} = 4\] which is just $ \frac {1}{a} + \frac {1}{b}\geq \frac {4}{a + b}$ for $ a,b > 0$.
28.06.2009 05:58
See also here: http://www.artofproblemsolving.com/Forum/viewtopic.php?t=102924
30.07.2009 11:39
i use cauchy ineequality at first let $ \frac{a-b}{b+c}+\frac{b-c}{c+d}+\frac{c-d}{d+a}+\frac{d-a}{a+b} =k$ $ k.((a-b)(b+c)+(b-c)(c+d)+(c-d)(d+a)+(d-a)(a+b)) \geq 0$ if a=1 b=2 c=3 d=4 then $ ((a-b)(b+c)+(b-c)(c+d)+(c-d)(d+a)+(d-a)(a+b))<0$then k<0 !!!!![/quote]
30.07.2009 11:57
table1388 wrote: i use cauchy ineequality at first let $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} = k$ $ k.((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) \geq 0$ if a=1 b=2 c=3 d=4 then $ ((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) < 0$then k<0 !!!!! [/quote] Cauchy requires positive, or at least non-negative, numbers.
30.07.2009 13:48
Mathias_DK wrote: table1388 wrote: i use cauchy ineequality at first let $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} = k$ $ k.((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) \geq 0$ if a=1 b=2 c=3 d=4 then $ ((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) < 0$then k<0 !!!!! Cauchy requires positive, or at least non-negative, numbers.[/quote] i dont think so,i studied my book again arent you wrong ?....
30.07.2009 14:08
table1388 wrote: Mathias_DK wrote: table1388 wrote: i use cauchy ineequality at first let $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} = k$ $ k.((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) \geq 0$ if a=1 b=2 c=3 d=4 then $ ((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) < 0$then k<0 !!!!! Cauchy requires positive, or at least non-negative, numbers. i dont think so,i studied my book again arent you wrong ?.... I'm sure You just found a counter example, when they were negative, didn't you?
30.07.2009 16:05
Mathias_DK wrote: table1388 wrote: Mathias_DK wrote: table1388 wrote: i use cauchy ineequality at first let $ \frac {a - b}{b + c} + \frac {b - c}{c + d} + \frac {c - d}{d + a} + \frac {d - a}{a + b} = k$ $ k.((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) \geq 0$ if a=1 b=2 c=3 d=4 then $ ((a - b)(b + c) + (b - c)(c + d) + (c - d)(d + a) + (d - a)(a + b)) < 0$then k<0 !!!!! Cauchy requires positive, or at least non-negative, numbers. i dont think so,i studied my book again arent you wrong ?.... Mathias_DK wrote: I'm sure You just found a counter example, when they were negative, didn't you? yes... .. you are right ...thanks.
11.09.2009 01:02
I think this problem is from Hung's book and limes123's solution is beautiful enough as given in the book.
08.01.2010 16:01
Very nice and easy inequality. This is problem equivalently to Baltic Way 1995.
Attachments:
123.doc (33kb)
08.01.2010 17:21
Zarif wrote: Very nice and easy inequality. This is problem equivalently to Baltic Way 1995. You just reposted a solution...
12.01.2010 07:31
I`m sorry Pain Rinnegan. I don`t know english and Latex!
13.11.2011 07:13
I think AG is enough the ineq is equivalent to $\sum\frac{a+c}{b+c}\ge 4$or $(a+b+c+d)(\frac{a+c}{(a+d)(b+c)}+\frac{b+d}{(a+b)(c+d)})\ge 4$ by AG,LHS$\ge 4(a+b+c+d)(\frac{a+c}{(a+d+b+c)^2}+\frac{b+d}{(a+b+c+d)^2})=4$.
10.01.2017 20:59
We can use trans inequality