n → ∞ lim ( n n n ! ) n 1 = ?
Bonus : Try converting the limit to a Riemann sum.
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.
Here is how I did it: I used the inequality e n 1 < n n n ! < e n n (I), took nth roots and then took the limit to obtain the answer, e 1 . To prove (I), take logs throughout to obtain − n < ∑ k = 1 n ln k − n ln n < ln n − n . Rearrange the terms to find ∑ k = 1 n − 1 ln k < n ln n − n < ∑ k = 1 n ln k . Now the middle term is 1 ∫ n ln x d x . The RHS and LHS are upper and lower Riemann sums for this integral, proving our claim.
Nice application of limit as a sum
really a very nice application of many concepts
The best solution I saw :D Keep up the good work
How do you know that you can express the improper integral ∫ 0 1 ln ( x ) d x as the limit of a Riemann sum? More specifically, how do you know that 0 ∫ 1 ln x d x = lim n → ∞ n 1 ∑ k = 1 n ln ( n k ) ?
Purely genius! I am totally impressed by your solution!
Beautiful question, and great solution !!
Smart exercise thank you for sharing like this exercise here
Beautiful Question!
How did you get the bounds of the integral?
We consider the limiting value of n k as k ranges over its possible values. At the beginning, k = 1 and n → ∞ . Hence, lim n → ∞ n k = 0 . This gives us the lower limit as 0.
Similarly, at the ending, k = n and n → ∞ . Hence, lim n → ∞ n k = 1 . This gives us the upper limit as 1. If k had ranged from 1 to 3 n , the upper limit would have been 3.
Ahhh. Got it. Thanks. :D
This can be done with Stolz-Cesaro or Stirling's but Stirling's is much more elegant.
From Stirling's formula we know that n ! ∼ 2 π n n n e − n So the limit can be written as n → ∞ lim ( n n e n 2 π n n n ) 1 / n And simplifying down we get n → ∞ lim e − 1 ( 2 π ) 1 / 2 n n 1 / 2 n The product on the right will go to 1 which is not difficult to show. Thus the answer is e − 1
Nice working!
Well, first of all, nice question!
Let n n n ! n 1 = y
n 1 ( l n ( n n n ! ) ) = l n y
e n 1 ( l n ( n n n ! ) ) = y
Now we need to find l i m n → ∞ y
We can find the power till infinity and then evaluate with e.
Hence, we need to find
l i m n → ∞ n 1 ( l n ( n n n ! ) )
= l i m n → ∞ n ( l n ( n ! ) − l n ( n n )
Using L'Hopital's rule,
= l i m n → ∞ ( d n d ( l n ( n ! ) ) − ( l n ( n ) + 1 ) ) [ l n ( n n ) = n l n ( n ) , and then we can use product rule to evaluate its derivative]
= l i m n → ∞ ( d n d ( l n ( n ! ) ) − l i m n → ∞ ( l n n ) − 1
We know that
l i m n → ∞ ( l n n ) = ∞
and
l i m n → ∞ ( d n d ( l n ( n ! ) ) = ∞ (because l n ( n ! ) = l n ( n ) + l n ( n − 1 ) + l n ( n − 2 ) . . . . . l n ( 2 ) + l n ( 1 ) and so when we take its derivative and then take the limit to infinity, the limit will go to ∞ )*
As a result,
= ∞ − ∞ − 1
= − 1
Now, don't forget that this is the power of e and the whole limit will be -
= e − 1
This solution is wrong. You can't perform arithmetic on infinities nor could you perform arithmetic on two divergent series.
Underrated, I think!
Hey Kartik, thanks! I find myself puzzling over the last few steps. Is it legal to subtract infinity from infinity like that?
The limit of d n d ln n ! as n goes to infinity is indeed infinity (by applying the limit-of-infinite-sum-is-an-integral principle) and the limit of ln n as n approaches infinity is also infinity, but are they the same infinity?
I seem to recall that ∞ − ∞ is an indeterminate form and cannot be evaluated. Perhaps the two limits that approach infinity should be combined and then evaluated to avoid this?
Could you explain how the question of the indeterminate form can be resolved? The solution I posted avoids this by splitting the factorial in the initial step itself instead of later.
Its defined on infinity
how can you substract infinite - infinite?
Problem Loading...
Note Loading...
Set Loading...
y = n → ∞ lim ( n n n ! ) n 1
ln y = n → ∞ lim n 1 ln ( n × n × ⋯ × n 1 × 2 × ⋯ × n )
ln y = n → ∞ lim n 1 ( ln n 1 + ln n 2 + ⋯ + ln n n )
ln y = n → ∞ lim n 1 k = 1 ∑ n ln n k
We know that such an infinite summation can be converted into an integral by the fundamental definition of integration as a limit of an infinite sum:
n → ∞ lim n 1 k = 1 ∑ n f ( n k ) = ∫ 0 1 f ( x ) d x
x = n k ⇒ d x = n 1
Hence,
ln y = ∫ 0 1 ln x d x = [ x ln x − x ] 0 1 = − 1
Note that substituting the lower limit requires us to evaluate the following:
x → 0 lim x ln x = 0
⟹ y = e − 1