Induction with more than one variable

µ³
Messages
68
Reaction score
0
Suppose we have a statement (say an equation) P(n, m), and we want to prove it works for arbitrary n and m.

Either

(1) It is sufficient to define Q(n) = P(n,m) or R(m) = P(n,m) and prove either Q(n) or R(m) by induction

(2) It is necessary to prove that P(n,m) => P(n+1,m) and P(n, m) => P(n, m+1), in addition to the base cases, to show that it works for arbitrary n and m.

Which one is correct !?
 
Physics news on Phys.org
µ³ said:
Suppose we have a statement (say an equation) P(n, m), and we want to prove it works for arbitrary n and m.

Either

(1) It is sufficient to define Q(n) = P(n,m) or R(m) = P(n,m) and prove either Q(n) or R(m) by induction

(2) It is necessary to prove that P(n,m) => P(n+1,m) and P(n, m) => P(n, m+1), in addition to the base cases, to show that it works for arbitrary n and m.

Which one is correct !?
If in (1) you prove Q(n) for ALL n or prove R(n) for ALL m, that would be sufficient. That would, in most cases involve proving P(n,m)=> P(n+1, m) to prove R(m) for a specific m, and then do another induction, proving R(m)=> R(m+1) to prove it true for ALL m. In other words, the two are really the same!
 
HallsofIvy said:
If in (1) you prove Q(n) for ALL n or prove R(n) for ALL m, that would be sufficient. That would, in most cases involve proving P(n,m)=> P(n+1, m) to prove R(m) for a specific m, and then do another induction, proving R(m)=> R(m+1) to prove it true for ALL m. In other words, the two are really the same!
The thing that confuses me is that for example to prove the binomial theorem, which is a statement P(x,y,n), we can simply prove that P(x,y,n) => P(x,y, n+1) and we don't have to prove it for all x and y, as we can just leave them arbitrary. Similarly why can't we prove, P(n,m) by proving P(n,m) => P(n, m+1), leaving n arbitrary and then assuming since we left n arbitrary that it will work for all n?
 
µ³ said:
The thing that confuses me is that for example to prove the binomial theorem, which is a statement P(x,y,n), we can simply prove that P(x,y,n) => P(x,y, n+1) and we don't have to prove it for all x and y, as we can just leave them arbitrary. Similarly why can't we prove, P(n,m) by proving P(n,m) => P(n, m+1), leaving n arbitrary and then assuming since we left n arbitrary that it will work for all n?
??But x and y are not integers so you wouldn't use "induction" on them anyway! Leaving x and y arbitrary- so that they can be any real numbers- does prove it for all x and y.
 
So what if P(n, 1) is trivially true for arbitrary n, then surely we can prove P(n, m) by showing P(n, m) => P(n, m+1) without having to do induction on n?
 
IF you can prove it for arbitrary n, yes.
 
Namaste & G'day Postulate: A strongly-knit team wins on average over a less knit one Fundamentals: - Two teams face off with 4 players each - A polo team consists of players that each have assigned to them a measure of their ability (called a "Handicap" - 10 is highest, -2 lowest) I attempted to measure close-knitness of a team in terms of standard deviation (SD) of handicaps of the players. Failure: It turns out that, more often than, a team with a higher SD wins. In my language, that...
Hi all, I've been a roulette player for more than 10 years (although I took time off here and there) and it's only now that I'm trying to understand the physics of the game. Basically my strategy in roulette is to divide the wheel roughly into two halves (let's call them A and B). My theory is that in roulette there will invariably be variance. In other words, if A comes up 5 times in a row, B will be due to come up soon. However I have been proven wrong many times, and I have seen some...
Back
Top