official quant thread for cat08 - Page 1295
PaGaLGuY.com - The Everything of MBA, CAT 2008, GMAT, XAT, IIM
         Home          MBA Forums         PG Office Blog         Contact Us         About Us                  Jobs @ PG
Exclusive Bschool Content:      Interviews      B-School Watch     MBAs speak     Placements     GMAT & MBA Abroad      Form Notifications
» Sponsors






Go Back   PaGaLGuY.com - The Everything of MBA, CAT 2008, GMAT, XAT, IIM > Exam Resources > Quantitative Questions and Answers

Notices
Quantitative Questions and Answers Discuss Quantitative and other Math related questions. Post your math doubts and get it solved by the smartest brains this side of the universe !

Tags: , ,

Reply
 
Thread Tools Display Modes
Re: official quant thread for cat08
Old
  (#12941)
ikruz
has no status.
Trainee PaGaL
 
Status: Offline
Posts: 54
Thanks: 97
Thanked 19 Times in 11 Posts
Join Date: Apr 2007
Location: Bangalore
Re: official quant thread for cat08 - 11-10-2008, 01:53 AM

Quote:
Originally Posted by ahaprashant View Post
Not exactly trial n error,
Look at the terms in sum form: (3x + 4y)
Look at the terms in prod form: (x^2, y^3)
Looking at these, can we say we need to think of some X-term, two of which when multiplied offer an x^2 term with certain coefficient, and when summed up offer 3x as the sum, which suggests it shud be 3x/2. Similarly for Y case.

Hope that helps. (Btw its prashant )


Pagal Padha.
Thanks a lot prashant.....It definitely helped.Its very clear for me now...

I have posted my doubt to similar question above. to maximise 2x^2 + 5y^2 + 8z^2 given 2x+3y+4z=100.

how will you approach that?.
   
Reply With Quote
Sponsored Links
Re: official quant thread for cat08
Old
  (#12942)
prakharc
has no status.
Hardcore PaGaL
 
prakharc's Avatar
 
Status: Offline
Posts: 642
Thanks: 139
Thanked 266 Times in 156 Posts
Join Date: Jul 2007
Location: Noida/Gurgaon
Age: 23
Re: official quant thread for cat08 - 11-10-2008, 08:24 AM

Quote:
Originally Posted by ikruz View Post
The question is to find out maximum value of 2x^2 + 5y^2 + 8z^2 ,if 2x+3y+4z=100.




Answer is indeed 4534.

Is trail and error the only possible way for handling this question.

The answer given by TIME is as follows -

For maximum value of the expression is when y is maximised because the ratio of the squares of the coeffs of x,y,z in 2x+3y+4z=100 is 4:9:16 whereas the ratio of coeffs of x^2,y^2, z^2 in 2x^2 + 5y^2 + 8z^2 is 4:10:16. Therefore y^2 has relatively more weightage.

But i could nt understand this logic as well..
Can any one throw some light.

Thanks..
They have done what I did......calculated the weightage to find out which one to maximize.....

What I did wasn't hit n trial.....I calculated that we shd maximize y and then did that.......I didn't try several values of x,y,z to get results


“The question isn't who is going to let me; it's who is going to stop me.”

My Mock CAT Scores AT:-

http://www.pagalguy.com/forum/1194247-post182.html
   
Reply With Quote
The Following User Says Thank You to prakharc For This Useful Post:
ikruz (11-10-2008)
Re: official quant thread for cat08
Old
  (#12943)
ikruz
has no status.
Trainee PaGaL
 
Status: Offline
Posts: 54
Thanks: 97
Thanked 19 Times in 11 Posts
Join Date: Apr 2007
Location: Bangalore
Re: official quant thread for cat08 - 11-10-2008, 10:51 AM

Quote:
Originally Posted by prakharc View Post
They have done what I did......calculated the weightage to find out which one to maximize.....

What I did wasn't hit n trial.....I calculated that we shd maximize y and then did that.......I didn't try several values of x,y,z to get results
Hi prakhar.

Sorry. i am not able to follow your logic.Could you explain the approach a bit more.

my doubt is -
1)what relation does the ratio of squares of the coeff of
x,y,z in 2x+3y+4z=100 have with the ratio of coeffs of x,y,z in 2x^2 + 5y^2 + 8z^2.?..

2) from this relation how do we conclude that y has to maximised?.

In your approach you have tried to tried 3 possible cases
a) x,y=0 z maximum
b) x,z=0 y maximum
c) y,z =0 x maximum.

and then concluded b) is the best option..am i right?.

Thanks for your help friend.
   
Reply With Quote
Re: official quant thread for cat08
Old
  (#12944)
prakharc
has no status.
Hardcore PaGaL
 
prakharc's Avatar
 
Status: Offline
Posts: 642
Thanks: 139
Thanked 266 Times in 156 Posts
Join Date: Jul 2007
Location: Noida/Gurgaon
Age: 23
Re: official quant thread for cat08 - 11-10-2008, 11:15 AM

Quote:
Originally Posted by ikruz View Post
Hi prakhar.

Sorry. i am not able to follow your logic.Could you explain the approach a bit more.

my doubt is -
1)what relation does the ratio of squares of the coeff of
x,y,z in 2x+3y+4z=100 have with the ratio of coeffs of x,y,z in 2x^2 + 5y^2 + 8z^2.?..

2) from this relation how do we conclude that y has to maximised?.

In your approach you have tried to tried 3 possible cases
a) x,y=0 z maximum
b) x,z=0 y maximum
c) y,z =0 x maximum.

and then concluded b) is the best option..am i right?.

Thanks for your help friend.
See you have to find max where we have square of x,y,z ...... so we have to maximize either x,y,z to get max value.

Now sum that we have to maximize is not symmetric in x,y,z......so we have to find relative importance of each term so as to find out which is most significant

So let's first maximize x.... x=50,y=0,z=0...then y=33.33,x=0,z=0...then z=25,x=0,y=0....this way we can see maximizing which term gives us highest sum => which term has the highest weightage in sum

We found that to be y...so we maximize y


what time people did was sot of like this only
2x+3y+4z=100...now if we take y,z=0 weightage of x^2=4..similarly for y and z...9,16 respectively

So in condition we have weightage of x^2,y^2,z^2 as 4:9:16

In sum we have 4:10:16.....so x and z remain same...but 9 of y in condition is giving 10 of y in sum....so if we maximize y we will get higher sum


“The question isn't who is going to let me; it's who is going to stop me.”

My Mock CAT Scores AT:-

http://www.pagalguy.com/forum/1194247-post182.html
   
Reply With Quote
The Following User Says Thank You to prakharc For This Useful Post:
ikruz (11-10-2008)
Re: official quant thread for cat08
Old
  (#12945)
ikruz
has no status.
Trainee PaGaL
 
Status: Offline
Posts: 54
Thanks: 97
Thanked 19 Times in 11 Posts
Join Date: Apr 2007
Location: Bangalore
Re: official quant thread for cat08 - 11-10-2008, 11:23 AM

question)The equation 2ax^2 + 3bx + 8 = 0 has equal roots when a and b are real numbers. Find the maximum value of a+ 2b.

a) -36/5 b) -22/3 c) -7 d) -64/9

How do we go about solving this problem

my first thought was that i i shud try to arrive at (a)(b^2) = k so that i can put a=b=b ..to get the minimum value of a+2b.

But here i can see 9b^2 = 64a.
from this i cant arrive at (a)(b^2) = k .So i am stuck here.

How to proceed?.
   
Reply With Quote
Re: official quant thread for cat08
Old
  (#12946)
naiquevin
freakshow excess!
Hardcore PaGaL
 
naiquevin's Avatar
 
Status: Offline
Posts: 372
Thanks: 68
Thanked 54 Times in 43 Posts
Join Date: Jul 2008
Location: Mumbai
Re: official quant thread for cat08 - 11-10-2008, 11:30 AM

find out maximum value of 2x^2 + 5y^2 + 8z^2 ,if 2x+3y+4z=100

consider 2x +3y + 4z = 100

had the question been to find maximum value of 4x^2 + 9y^2 + 16z^2 which is same as (2x)^2 + (3y)^2 + (4z)^2........(I)
then we would have maximised any one of 2x, 3y or 4z and proceeded
to find the maximum value

But we are asked to maximise 2x^2 + 5y^2 + 8z^2.......(II)
The ratio of coefficients is 4:10:16

whereas the ratio of coeffcients in exprn I is 4:9:16

if x and z are maximised then the increase in value of II over I will not be as much as when y is maximised


My blog to share some guitar gyan!
http://guitargonemad.blogspot.com/

I've been imitated so well I've heard people copy my mistakes
---Jimi 'THE IMMORTAL' Hendrix
   
Reply With Quote
The Following User Says Thank You to naiquevin For This Useful Post:
ikruz (11-10-2008)
Re: official quant thread for cat08
Old
  (#12947)
ikruz
has no status.
Trainee PaGaL
 
Status: Offline
Posts: 54
Thanks: 97
Thanked 19 Times in 11 Posts
Join Date: Apr 2007
Location: Bangalore
Re: official quant thread for cat08 - 11-10-2008, 11:47 AM

thanks naiquevin and prakhar..you are people are so helpful, i got the concept now!!...

thanks for ur time.
   
Reply With Quote
Re: official quant thread for cat08
Old
  (#12948)
masoom
Searching for Kismat Connection!!!
Addicted PaGaL
 
masoom's Avatar
 
Status: Offline
Posts: 1,037
Thanks: 103
Thanked 255 Times in 184 Posts
Join Date: Jul 2008
Location: Gurgaon
Age: 23
Re: official quant thread for cat08 - 11-10-2008, 12:31 PM

a graph is given as attcahed...

using this sketch the graph of y vs x...
Attached Files
File Type: doc y.doc (26.0 KB, 14 views)


moi Blog :
http://masoom.livejournal.com/

" prabisi nagar keeje sab kaaja...Hriday rakhi kosalpur raja "

-- "Ram Charit Manas"
   
Reply With Quote
Re: official quant thread for cat08
Old
  (#12949)
Vishal2704
Dejected...
Trainee PaGaL
 
Status: Offline
Posts: 96
Thanks: 8
Thanked 30 Times in 21 Posts
Join Date: Jan 2007
Location: Hyderabad
Age: 25
Re: official quant thread for cat08 - 11-10-2008, 12:45 PM

Quote:
Originally Posted by masoom View Post
a graph is given as attcahed...

using this sketch the graph of y vs x...
y-x = k(y+x) k>1

y/x = (k+1)/1-k

hence slope of y to x is -ve as 1-k <0
Attached Files
File Type: doc ans.doc (26.0 KB, 11 views)
   
Reply With Quote
Re: official quant thread for cat08
Old
  (#12950)
MaskedMenace
-- schizophrenia --
Addicted PaGaL
 
MaskedMenace's Avatar
 
Status: Offline
Posts: 897
Thanks: 176
Thanked 155 Times in 127 Posts
Join Date: Aug 2008
Location: Hyderabad
Age: 21
Re: official quant thread for cat08 - 11-10-2008, 12:52 PM

Quote:
Originally Posted by masoom View Post
a graph is given as attcahed...

using this sketch the graph of y vs x...
v get y=[(1+k)/(1-k)]x line for k>1


Menace!!
   
Reply With Quote
Reply


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off
Forum Jump

» Sponsors










PaGaLGuY.com is not responsible for the views and opinions of the posters.
PaGaLGuY.com is an Inzane Labs Private Limited production.
Hosted on servers powered by Neutral Web