Let k ∈ ( 0 , 1 ) , and let the sequence ( B n ) n = 0 ∞ be defined by:
B 0 = k , B 1 = k 2 and B n + 2 = k B n + 1 + k 2 B n for integers n ≥ 0 .
If the value of the expression n = 0 ∑ ∞ n + 1 B n can be expressed as:
γ 1 ln ( 1 − α k 1 − β k )
where γ ∈ Z ; α , β ∈ R and α > 0 , β < 0 . Find the value of ⌊ α + β + γ ⌋ .
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.
We can also show the solution of recurrence relation by "formally finding linear recurrence solutions" by using characteristic equation method.
Log in to reply
Yup! Post your solution!
This is actually a much easier method.
Log in to reply
Checking by direct substitution and using generalized formulas is not an easy method, at least for me. You never know what problem you're getting tomorrow(not a fibonacci maybe).
Well, actually the series converges if: − ϕ ≤ k < ϕ .
We can show B n = k n + 1 F n by induction as well.
Log in to reply
Yes, then you should change the statement to − ϕ < k < ϕ .
Log in to reply
Well, I can. But it doesn't matter as (0,1) lies within the valid range.
Log in to reply
@Satyajit Mohanty – I think you have a confusion here. From the ratio test, the seies converges if ∣ k ∣ < 1 / ϕ , which definitely does not contain ( 0 , 1 ) . Check again.
Problem Loading...
Note Loading...
Set Loading...
It is clear that B n = k n + 1 F n (check by direct substitution!), where F n is the n th Fibonacci number. Now it is well-known that the Generating function of the Fibonacci series is given by : n = 0 ∑ ∞ F n z n = 1 − z − z 2 1 within the radius of convergence ∣ z ∣ < ϕ 1 . Hence for all ∣ k z ∣ < ϕ 1 , we have : n = 0 ∑ ∞ B n z n = k n = 0 ∑ ∞ F n ( k z ) n = 1 − k z − k 2 z 2 k Take k < ϕ 1 and hence the above series is uniformly convergent in the interval ( 0 , 1 ) . Thus we can integrate both sides of the power series to yield n = 0 ∑ ∞ n + 1 B n = ∫ 0 k 1 − x − x 2 1 d x = 5 1 lo g ( 1 − k ( 5 + 1 ) / 2 1 + k ( 5 − 1 ) / 2 ) which yields the result.
Note : I doubt whether the series is convergent for k > ϕ 1 . Hence the statement is the problem needs to be modified by restricting 0 < k < ϕ 1 , where ϕ is the Golden ratio.