The Fibonacci sequence can be defined with induction as F n = F n − 1 + F n − 2 , where F 1 = 0 and F 2 = 1 . It is also well-known that the limit of the ratio between two consecutive terms lim F n F n + 1 is the golden ratio ϕ .
In the generalized Fibonacci sequence below, the original Fibonacci sequence is F n 2 . F n m = i = 1 ∑ m F n − i m with F n m = 0 if n < m and F m m = 1 . Every sequence will also have a generalized golden ratio ϕ m = n → ∞ lim F n m F n + 1 m .
What is the limit of the generalized golden ratio below? m → ∞ lim ϕ m = m → ∞ lim n → ∞ lim F n m F n + 1 m
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Probably the simplest solution. Start with m being the infinite sum and you've proved it's 2.
If there's one weakness it's finding 2 by observation but that's just the purist in me speaking. The empiricist says it's fine.
how to see the second equality in the last line?
Log in to reply
lim m → ∞ ϕ m = lim m → ∞ lim n → ∞ F n m F n + 1 m is given in the problem
By the definition of the sequence, we have F n m + F n + 1 m + F n + 2 m + ⋯ F n + m − 1 m = F n + m m For large enough n , this becomes 1 + ϕ m + ϕ m 2 + ⋯ ϕ m m − 1 = ϕ m m ⟹ ϕ m − 1 ϕ m m − 1 = ϕ m m ⟹ ϕ m m ( ϕ m − 2 ) = − 1 As m goes to infinity, ϕ m m goes to infinity, so in order for the product to converge, ϕ m − 2 must tend to 0, therefore the answer is 2.
Assuming ϕ m exists for every m ...
We can divide both sides of F n m = i = 1 ∑ m F n − i m by F n − m m and take the limit as n → ∞ to find ϕ m m = i = 1 ∑ m ϕ m m − i = ϕ m − 1 ϕ m m − 1 ⟹ ϕ m m ( ϕ m − 2 ) + 1 = 0 so ϕ m is a root of x m ( x − 2 ) + 1 = 0 .
Since for each m , the sequence n ↦ F n m is increasing, we must have ϕ m ≥ 1 . Further, if we look at the first equation I gave for ϕ m , we can see that ϕ m = 1 ⟹ 1 = ϕ m m = i = 1 ∑ m ϕ m m − i = m so, since we're only interested in lim m → ∞ ϕ m , we may assume ϕ m > 1 . Further, since for x ≥ 2 , we have x m ( x − 2 ) + 1 ≥ 1 > 0 , we also have ϕ m < 2 for all m .
So far, we have shown that if m > 1 , then ϕ m ∈ ( 1 , 2 ) is a root of f m ( x ) = x m ( x − 2 ) + 1 . To strengthen this, note that x = 1 is a root of f m and f m ′ ( x ) = ( m + 1 ) x m − 2 m x m − 1 = x m − 1 ( ( m + 1 ) x − 2 m ) has only one root x = m + 1 2 m ∈ ( 1 , 2 ) , so by Rolle's theorem (or more generally the Mean Value Theorem) we can conclude that ϕ m is the unique root ∈ ( 1 , 2 ) of f m .
To finish the proof, we note that for any 0 < ε < 1 , m > − lo g 2 − ε ( ε ) ⟹ f m ( 2 − ε ) = ( 2 − ε ) m ( − ε ) + 1 < 0 ⟹ f m has a root ∈ ( 2 − ε , 2 ) ⊂ ( 1 , 2 ) ⟹ 2 − ε < ϕ m < 2 (Intermediate Value theorem) (Since ϕ m is the unique root ∈ ( 1 , 2 ) )
Therefore, we have shown by the formal definition of the limit that m → ∞ lim ϕ m = 2
It's perhaps a little confusing when the question uses superscript m is both as an exponent and to denote the m-th series.
Log in to reply
Personally, I would have used a "double subscript" on the sequence if I had designed it, but some people hate multiple subscripts on principle.
For instance, most research mathematicians will have at some point seen someone try to use five or more subscripts on the same term in a presentation, and it just becomes ridiculous after a while.
I'm going to play a little fast and loose with limits towards infinity here.
Extend this generalized sequence to a very high value of m . The first terms are a bunch of zeroes then the one but then comes [ . . . 0 , 0 , 0 , 1 , ] 1 , 2 , 4 , 8 , 1 6 , 3 2 , . . . . This doubling sequence will occur for m terms. If we make m → ∞ then the ratio between successive terms is 2 .
The generating function for the series is F ( z ) = z + z F ( z ) + z 2 F ( Z ) + . . . + z m F ( z ) rearranging the terms ∴ F ( z ) = 1 − i = 1 ∑ m z i z
as m → ∞ , m → ∞ lim i = 0 ∑ m z i = 1 − z 1 ∴ F ( z ) = 2 z − 1 z 2 − z ∴ F ( z ) = 2 z − 4 1 + 4 1 ⋅ 1 − 2 z 1 ∴ F ( z ) = 2 z − 4 1 + 4 1 ⋅ i = 0 ∑ ∞ ( 2 z ) i ∴ n → ∞ lim ϕ m = 2
Since the text gives us that F n m = i = 1 ∑ m F n − i m we can evaluate F n + 1 m as follows: F n + 1 m = i = 1 ∑ m F n + 1 − i m = F n m + F n − 1 m + … + F n − m + 1 m = F n m + F n − 1 m + … + F n − m + 1 m + F n − m m - F n − m m = 2 ⋅ F n m - F n − m m . It's clear that F n m F n + 1 m = F n m 2 ⋅ F n m − F n − m m = 2 -- F n m F n − m m . Since Fibonacci - like sequences, obviously, grow monotonically, for n sufficient large (i.e. for n approaching infinity) the fraction becomes 0 and the result we find is 2
For F m , each term (except for the first m ones) is equal to the sum of m previous terms. So if we expand two adjacent terms F k m and F k + 1 m we get:
F k m = Σ i = k − m k − 1 F i m and
F k + 1 m = Σ i = k − m + 1 k F i m .
Using the overlap in the ranges: F k + 1 m = Σ i = k − m k − 1 F i m − F k − m m + F k m = 2 F k m − F k − m m .
The ratio between adjacent terms then is F k m F k + 1 m = 2 − F k m F k − m m .
We are told that ϕ m = lim k → ∞ F k m F k + 1 m exists. From this we can infer that lim k → ∞ F k m F k − m m = ϕ m − m . And since for m>1, ϕ m must be larger than 1(*), we know lim m → ∞ ϕ m − m = 0 , and lim m → ∞ ϕ m = lim m → ∞ lim k → ∞ F k m F k + 1 m = 2 − 0 .
(*) [Having ϕ m = 1 would lead to a contradiction: if m terms are (limitwise) equal, adding them up gives a next term m times as large.]
Let m ∈ N You can find that
F n m F n + 1 m = ∑ i = 1 m F n m F n + 1 − i m = ∑ i = 0 m − 1 F n m F n − i m
and when you can take the limit
lim n → ∞ F n m F n + 1 m = lim n → ∞ ∑ i = 0 m − 1 F n m F n − i m = ∑ i = 0 m − 1 lim n → ∞ F n m F n − i m = ∑ i = 0 m − 1 ϕ m − i = ∑ i = 0 m − 1 ( ϕ m 1 ) i ,
which is a geometric series, so
ϕ m = lim n → ∞ F n m F n + 1 m = 1 − ϕ m 1 1 − ( ϕ m 1 ) m .
You can then rearrange it to become.
ϕ m ( 1 − ϕ m 1 ) = 1 − ( ϕ m 1 ) m ⇒ ϕ m − 1 = 1 − ( ϕ m 1 ) m ⇒ ϕ m = 2 + ( ϕ m 1 ) m
Because ϕ m 1 ≥ 0 it follows that ϕ m ≥ 2 ∀ m ∈ N
Therefore are lim m → ∞ ϕ m ≥ 2
You can also see that
lim m → ∞ ϕ m = lim m → ∞ 2 + ( ϕ m 1 ) m ≤ 2 + ( 2 1 ) m = 2
Therefore is
lim m → ∞ ϕ m = 2
n-nacci sequence has this ratio as the root of x+x^(-n)=2, which is nearest to 2 in the large-n limit. Answer=2.
As shown elsewhere each solution of the m-th sequence can be expressed as phi(m)^m=sum of all lower powers of phi(m).
For m=2, the solution is 1.62 >1.
We can demonstrate that this is an increasing function w.r.t m by multiplying both sides of the equation by phi (m).
Phi (m) ^(m+1) = Sum of lower powers of phi (m) plus 1. Which with a simple derivative allows us to demonstrate that we are dealing with a monotonic increasing function.
We know from the simple solution shown earlier that phi(m)^(m-1)*(phi(m)-2) = -1 and dividing by phi(m)^(m-1) we see that Phi(m)-2 is a very small negative number which approaches 0 at the limit
42 is the answer of the sense of life, so 4 divided by 2 is 2. that was realy easy
float f(int n,int m) {int i,s=0; if(n<m) return 0; if(n==m) return 1; for(i=1;i<=m;i++) { s=s+f(n-i,m);
}
return s;
}
int main() {int m,n; float g;
scanf("%d,%d",&m,&n);
g=f(n+1,m)/f(n,m); printf("%.10f",g); } JUST WRITE THIS C PROGRAMME, AND INPUT SOME VALUE LIKE (60,70) FOR n AND m , you will find the ratio 2
Problem Loading...
Note Loading...
Set Loading...
As lim m → ∞ , the next number in the sequence would be the sum of all the numbers before it, so the first non-zero numbers of F are 1 , then 0 + 1 = 1 , then 0 + 1 + 1 = 2 , then 0 + 1 + 1 + 2 = 4 , then 0 + 1 + 1 + 2 + 4 = 8 , and so on, for the sequence 1 , 1 , 2 , 4 , 8 , …
After the first 1 , we observe that the numbers after it are successive powers of 2 . In other words, if G p is the sequence of F n starting at the second 1 so that G 0 = 1 , G 1 = 2 , G 2 = 4 , and so on, then G p = 2 p . This observation can be proved inductively: G 0 = 2 0 = 1 , and assuming G p = 2 p , G p + 1 = 1 + k = 0 ∑ p 2 k = 1 + 1 − 2 1 − 2 p + 1 = 2 p + 1 . Therefore,
m → ∞ lim ϕ m = m → ∞ lim n → ∞ lim F n m F n + 1 m = p → ∞ lim G p G p + 1 = p → ∞ lim 2 p 2 p + 1 = 2