Solve Summation Question: T(n) Recurrence Form

  • Context: Graduate 
  • Thread starter Thread starter jamesvanboxtel
  • Start date Start date
  • Tags Tags
    Summation
Click For Summary

Discussion Overview

The discussion revolves around a recurrence relation T(n) involving a summation, with participants exploring methods to simplify or find a closed form for the expression. The scope includes mathematical reasoning and potential applications of recurrence relations.

Discussion Character

  • Exploratory, Mathematical reasoning, Debate/contested

Main Points Raised

  • One participant presents the recurrence relation T(n) and requests assistance in simplifying it or finding a closed form.
  • Another participant suggests manipulating the recurrence by moving terms and summing over n, indicating the need to consider separate cases for odd and even n.
  • A different participant notes that the constant term can be factored out of the summation and proposes an alternative approach of summing terms in pairs to simplify the analysis.
  • A later reply mentions discovering a pattern by entering results into a sequence database, indicating progress in understanding the recurrence.

Areas of Agreement / Disagreement

Participants express various approaches and ideas, but there is no consensus on a single method or solution. The discussion remains unresolved with multiple competing views on how to tackle the recurrence.

Contextual Notes

Participants highlight the need to consider cases for odd and even n, and there are indications of potential complexities in the summation that remain unaddressed.

jamesvanboxtel
Messages
2
Reaction score
0
I have the following recurrence that I am trying to come up with atleast a simplified version if not a closed form.

[tex]T(n) = T(n-1) + \sum_{i=1}^{(n-1)/2} [(n-(i+1)) * (i-1) * 2 + 2][/tex]

in addition if n is even I must add the following to T(n)
[tex]((n/2) - 1)^2[/tex]If any of you can help that would be awesome.

BTW this forum looks really cool.
 
Last edited:
Physics news on Phys.org
This isn't all that clear but why not move the term on the right and have T(n) - T(n-1)
then sum over n from 1 to n (if summing over n with top limit n is confusing index to a different letter, it won't make a difference in the end)on both sides for a double sum on the right.
Then the left will be have T(n) - T(0) , the T(0) you may know.
As for the right since only multiplication seems to be involved prominently the original summational need not be a problem though I'm not sure of the second that I suggested(over n)
You'll need to consider separate cases for n odd or even also. Hope I interpreted correctly.
 
Of course, the constant +2 can be taken out of the sum; summing +2 in a loop, 'm' times, is just adding 2*m.

One question: do you really need each and all T(n)? Because maybe you could sum the terms in pairs, that is, construct a new series 'A' where A(0)=T(0)+T(1), A(1)=T(2)+T(3), A(2)=T(4)+T(5) and so on, so that the special case for even/odd terms dissapears.
 
Turns out I was able to enter the first few results here and come up with what the pattern was.

http://www.research.att.com/~njas/sequences/

Thanks for trying though guys.
 
Last edited by a moderator:

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 4 ·
Replies
4
Views
3K
Replies
3
Views
2K
  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 16 ·
Replies
16
Views
3K
Replies
5
Views
3K