| Author |
Message |
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 14:30:25 IST
|
|
|
Lemme see who solves it first Determine range of [ a/(a+b+d) + b/(a+b+c) + c/(b+c+d) + d/(a+c+d) ] for positive reals a, b, c, d. This is very simple
|
|
|
|
|
|
|
|
<=4/3 ?
|
Nitwit Blubber Odment Tweak
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 14:59:49 IST
|
|
|
i think the min tends to 1 and max tends to 2??
i.e range is (1,2) 
|
JEE and OLYMPIA INFINATUM
http://iit-redefined.theforum.name/index.php
join the revolution!! |
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 15:00:22 IST
|
|
|
pl temme how it tend to 2
|
Nitwit Blubber Odment Tweak
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 15:12:26 IST
|
|
|
let a>b>c>d b+c+d < b + c+ a
add on both sides and the RHS will be 1 now, adding that on both sides, we get the inequality 
NOTE: i did this same sum very recently, which is y i told d ans immediately
|
JEE and OLYMPIA INFINATUM
http://iit-redefined.theforum.name/index.php
join the revolution!! |
this reply: 5 points
(with 1 
in 1 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 15:12:34 IST
|
|
|
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 15:19:37 IST
|
|
|
edited
but i think there shud be a stricter inequality
|
Nitwit Blubber Odment Tweak
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 17 Mar 2008 18:32:21 IST
|
|
|
I am not sure whether you are asking the range in terms of a,b,c and d or actual values. greater than 1 is easily proved. Also, Sorry, that should be a strict inequality. But if it is in terms of a,b,c and d, we must note that if a>b>c>d, then So, if we call the given expression S, Rearrangement Inequality gives, and 
|
this reply: 5 points
(with 1 
in 1 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 18 Mar 2008 09:45:22 IST
|
|
|
also, because of the homogoneity of the expression we can assume a+b+c+d = 1. I think smoothing principle would give you the minimum as 4/3. Could someone pls confirm this?
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 18 Mar 2008 10:05:04 IST
|
|
|
no that's not true S can be considered as a function in each variable seperately holding out others constant and in each case it is continiouas at each point in the reals - the point where it blows up also it's clear that 1<=S<=2 we now prove that one can get as close one wants to both of them set a=b=u and c=d=v now take the limit x--->1 and y--->0 along any path u want(if u wnat them to go depenedently otherwise no prob)(as the cauchy-reimann conditions hold true) we get the limit as 1 set a=c=u and b=d=v and take limits as u--->1 and v--->0if u wnat them to go depenedently otherwise no prob) we get limit as 2 so we can get arbitrarily close
|
this reply: 5 points
(with 1 
in 1 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
![[Post New]](/templates/default/images/icon_minipost_new.gif) 18 Mar 2008 23:36:19 IST
|
|
|
Ya thats rite Ans is 1<=A<=2
|
this reply: 0 points
(with 0 
in 0 votes ) [?]
|
|
You have to be logged on to rate
|
|
|
|
|