[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference rusure::math

Title:Mathematics at DEC
Moderator:RUSURE::EDP
Created:Mon Feb 03 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2083
Total number of notes:14613

1300.0. "Summation Problem" by CSC32::S_JOHNSON (Lifetime Member of Aye Phelta Thi) Sun Sep 23 1990 21:33

Hi,

I need help with a sumnation problem.  Can someone give me some help with
the following equation.  I need to show that it is true.

      n
     ---
     \     n-i  2       n  2
     /    2    i    <= 2  n 
    /----
     i=1

Thanks for any and all replies.

Scott Johnson
T.RTitleUserPersonal
Name
DateLines
1300.1SSDEVO::LARYunder the Big Western SkyMon Sep 24 1990 04:5026
>      n
>     ---
>     \     n-i  2       n  2
>     /    2    i    <= 2  n 
>    /----
>     i=1

I'm sure there's some closed-form expression for the sum, but you
can very simply can get a lot tighter bound than what you need:

the sum can be rewritten:

	 n-1      n-2      n-3       n-4
	2    + 4*2    + 9*2    + 16*2    + ...

Or, rearranging slightly:

	  n-1    n-3       n-2      n-3       n-4
	(2    + 2   ) + 4*2    + 8*2    + 16*2    + ...

			   n
All the terms shown are <=2 , and so are all the unshown terms since

(i+1)�/i� < 2 for i > 3
						       n
So it would appear that the sum is always less than n*2
1300.2by contradiction alsoSMAUG::ABBASIMon Sep 24 1990 05:2254
    by contradiction, Assume it is false, i.e
    

      n
     ---
     \     n-i  2       n  2
     /    2    i    > 2  n 
    /----
     i=1
    
    i.e
       
        n
       ---
 n     \          2       n  2
2      /         i     > 2  n 
      /        -----
     /           i
    /          2
   /----
     i=1
    
    i.e
    
      ---    2
     \     i         2
     /     ---    > n
    /       i
    --     2
    
                                    2         2
     or  (  1  +  4     +   9   +  n    )  > n
           ---   ---       ---    ---
            2    2^2       2^3    2^n
    
    
     or (  1  +   4     +   9   +  1    )  > 1
          ---   -----      ----   ---
            2    2  2       3  2   n
         2*n    2  n       2  n   2
    
    for n=1 ,  1/2 > 1 , not valid, hence assumption is wrong, hence proof
    for n=2..infinity  Left hand side goes to 0, hence 0>1 not valid, hence
    assumption is wrong, hence proof.
    
    p.s.
                              x=n
    you can also proof by     /     2
                             |    x
                             |   ---   dx  and show that this definite I is
                            /      x       <= n^2
                            x=1   2
    
    /naser
1300.3closed form for sum.CADSYS::COOPERTopher CooperMon Sep 24 1990 11:3221
RE: .1

>>      n
>>     ---
>>     \     n-i  2       n  2
>>     /    2    i    <= 2  n 
>>    /----
>>     i=1
>
>I'm sure there's some closed-form expression for the sum, ...

    With Maple's help:

      n
     ---
     \     n-i  2      n    2
     /    2    i  = 6*2  - n  - 4n - 6
    /----
     i=1

					    Topher
1300.4One more proof.CADSYS::COOPERTopher CooperMon Sep 24 1990 11:4619
    Oh yes, nearly forgot.  Previous postings adequately proved it, but
    using the closed form (and assuming n > 0).

	       n    2               n 2
	    6*2  - n  - 4n - 6 <=? 2 n

    Divide both sides by the (positive) left side:

	    6	   1	 4     6
	   --- - (--- + --- + ----) <=? 1
	     2      n    n     n 2
	    n      2    2 n   2 n

    The first term is < 1 for n>2.  The other terms can only serve to
    reduce the size of the lefthand side of the inequality and so can
    be ignored for n>2.  The relation is easily checked for n=1 and n=2,
    QED.

				    Topher
1300.5ThanksCSC32::S_JOHNSONLifetime Member of Aye Phelta ThiMon Sep 24 1990 12:393
    Thanks. I appreciate everyone's responses.
    
    scott
1300.6how did MAPLE find it ?SMAUG::ABBASIMon Sep 24 1990 14:3513
    ref .3
    
    >With Maple's help:

    >  n
    >  ---
    >  \     n-i  2      n    2
    >  /    2    i  = 6*2  - n  - 4n - 6
    > /----
       i=1
    
    Ok, assuming we did not have this amazing MAPLE, how would one go about
    deriving this by hand.
1300.7Guess the answer, then prove it by inductionDECWET::BISHOPNihon ga natsukashii ...Mon Sep 24 1990 15:177
re .6

You can compute it for several values of n and get the pattern from that, then
prove it by induction.  For a formula as complex as this one, however, finding
the pattern would be pretty hard.

Avery
1300.8thanks, But now Iam more curious abot MAPLE ?SMAUG::ABBASIMon Sep 24 1990 16:0012
    rec .7
    thanks Avery, 
>You can compute it for several values of n and get the pattern from that, then
>prove it by induction.  For a formula as complex as this one, however, finding
>the pattern would be pretty hard.
    
    I see how I may try it by hand, But then How did MAPLE do it ?
    what analytical method could it have used ? it could not have
    used the above method ?
    thanks
    
    
1300.9TRACE::GILBERTOwnership ObligatesMon Sep 24 1990 19:2033
   n
  ---         n+1
  \     i    x    - 1
  /    x  =  --------
  ----         x - 1
  i=0

Taking derivatives of both sides,

   n
  ---                   n    n+1
  \       i-1    (n+1) x    x    - 1
  /    i x    =  -------- - --------
  ----            x - 1     (x-1)^2
  i=1

Multiply both sides by x, and take derivatives again, 

   n
  ---                   2  n     n+1             n+1                n
  \     2  i-1     (n+1)  x     x   - 1    2 x (x   - 1)   2 (n+1) x  + 1
  /    i  x     =  ---------- - -------- + ------------- - --------------
  ----              x - 1       (x-1)^2       (x-1)^3         (x-1)^4
  i=1

Now multiply both sides by `x 2^n', and then let x = �.

   n
  ---
  \     2  n  -i       n    2
  /    i  2  2   =  6 2  - n  - 4 n - 6
  ----
  i=1
1300.10like itHERON::BUCHANANcombinatorial bomb squadTue Sep 25 1990 05:515
Re:           <<< Note 1300.9 by TRACE::GILBERT "Ownership Obligates" >>>

That's neat!

Andrew.
1300.11How did you get Maple to do it?CSC32::S_JOHNSONLifetime Member of Aye Phelta ThiTue Sep 25 1990 18:0316
    >With Maple's help:

    >  n
    >  ---
    >  \     n-i  2      n    2
    >  /    2    i  = 6*2  - n  - 4n - 6
    > /----
       i=1
    
    I just managed to get maple loaded onto one of our systems.  What did
    you enter to get the result?  I tried sum((2^(n-i))*(i^2)); and it does
    not look the same.
    
    thanks for any replies.
    
    scott
1300.12GUESS::DERAMODan D&#039;EramoTue Sep 25 1990 19:4618
>>    I just managed to get maple loaded onto one of our systems.  What did
>>    you enter to get the result?  I tried sum((2^(n-i))*(i^2)); and it does
>>    not look the same.

	You have to tell Maple what you are "summing over".  If you
	tell it to sum (2^(n-i))*(i^2) as i varies from 1 to 10 and
	n varies from 1 to 5, then it will compute those fifty terms
	and add them and tell you that number.  If you tell Maple to
	sum (2^(n-i))*(i^2) as n varies from -3i to 17i+12, it will
	attempt to sum it "symbolically", expressing its result as a
	function of i.  In this case you need to tell Maple that you
	are summing over i, as i varies from 1 to n, and to give the
	result symbolically as a function of n.  I don't know Maple
	syntax so I can't tell you how to do that; it might be some-
	thing like sum( (2^(n-i))*(i^2) , i , 1 , n ) or perhaps
	sum( (2^(n-i))*(i^2) ; i=1,...,n ).

	Dan
1300.13Try simplifying the resultALLVAX::JROTHIt&#039;s a bush recording...Tue Sep 25 1990 20:0520
>    I just managed to get maple loaded onto one of our systems.  What did
>    you enter to get the result?  I tried sum((2^(n-i))*(i^2)); and it does
>    not look the same.

say something like

	s := sum(2^(n-i)*i^2,i=1..n);

and then say

	simplify(s);

The online help for Maple is pretty good - I don't use Maple regularly
enough to be any kind of expert, but rarely have to look at the manual.

You just have to play a little.

I wish I had tools like this ages ago...  oh well.

- Jim
1300.14Delaunay's wishes the sameSMAUG::ABBASITue Sep 25 1990 20:3025
   ref .-1     <<< Note 1300.13 by ALLVAX::JROTH "It's a bush recording..." >>>
>I wish I had tools like this ages ago...  oh well.
>- Jim
    
    jim, this may makes you feel a little better.
    extracted from computing supplementum 4, computer algebra symbolic and
    algebric computation, 1982, springer-verlag.
    
    "the introduction of a new technique (the Lie transform) allowed Deprit
    to recompute the reduced Hamiltonian of the main problem in lunar
    theory. Delaunary spent TWENTY YEARS to the complete the calculation by
    HAND. Deprit used only 20 HOURS on a small PC using a computer algebra
    package"
    
    Delaunary lived in the 1860's .
    
    also the package found out that Delaunay results were correct up to 
    order 9.
    
    I bet the same calculation could be done today in 20 MINUTES.
    
    I felt sad for the Late Delaunary after reading this. 
    I wondar what he would say if he knew this !
    /naser