The Student Room Group

Summations

jkNx=N2,    j=1,k=1 \sum_{j \neq k}^{N} x = N^2, \ \ \ \ j=1,k=1

What is xx?

I thought it would be 1, but does this not give the answer N2NN^2-N?
Reply 1
Original post by adie_raz
jkNx=N2,    j=1,k=1 \sum_{j \neq k}^{N} x = N^2, \ \ \ \ j=1,k=1

What is xx?

I thought it would be 1, but does this not give the answer N2NN^2-N?

Yep, because of the jkj \neq k condition. Is x allowed to depend on N? If it is, then there's a simple tweak you can do (it helps me to visualise an NxN matrix, where the required quantity is the sum of all the entries, and where the diagonal entries are 0).
Reply 2
Original post by Smaug123
Yep, because of the jkj \neq k condition. Is x allowed to depend on N? If it is, then there's a simple tweak you can do (it helps me to visualise an NxN matrix, where the required quantity is the sum of all the entries, and where the diagonal entries are 0).


Yep, x can depend on N. Is x=N2N2N=NN1 x = \frac{N^2}{N^2-N} =\frac{N}{N-1} ??
(edited 11 years ago)
Reply 3
Original post by adie_raz
Yep, x can depend on N. Is x=N2N2N=NN1 x = \frac{N^2}{N^2-N} =\frac{N}{N-1} ??

Yep, exactly - then each row of the matrix adds up to N, and there are N rows, so the whole thing comes to N^2.
Reply 4
Original post by Smaug123
Yep, exactly - then each row of the matrix adds up to N, and there are N rows, so the whole thing comes to N^2.


Thanks, what a wonderfully easy way to think about it, I shall never have an issue like this again now!
Reply 5
Original post by adie_raz
Thanks, what a wonderfully easy way to think about it, I shall never have an issue like this again now!

No problem :smile: it doesn't work so well for things like the sum of 1/x, because it's not so clear what each row sums to, but for constants it can be a simpler way. (Makes use of the innate ability to picture things and work with concrete objects, I suppose, much like the way I do Bayes's Law with a kind of Carroll diagram.)

Quick Reply

Latest