3 ! 1 2 + 4 ! 2 2 + 5 ! 3 2 + …
If the above summation can be expressed as S , find ⌊ 1 0 0 0 S ⌋ .
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.
Is this the amount of rigor with which mathematicians write proofs? ( I mean checking the convergence and everything )
Log in to reply
This is nothing compared to some of the rigorous proofs I have seen in mathematical research papers. I'm not even sure whether my solution is rigorous or not.
Testing convergence of series is pretty standard when dealing with infinite series because otherwise we'd be performing arithmetic operations on divergent sums which can result in a lot of unexpected stuff.
Log in to reply
Your solution is indeed rigorous.
Because the question asked for the first 3 digits of S , one can cheat by simply finding the sum of the first few terms.
Let us use the expansion of the exponential function around zero, which is given by, for all x ∈ R e x = n ≥ 0 ∑ n ! x n Dividing both sides by x 2 , we find, for all x ∈ R / { 0 } x 2 e x = n ≥ 0 ∑ n ! x n − 2 . Then, as this sum is absolutely convergent, we can take the derivative with respect to x at both sides of the equation and interchange the derivative with the sum, as follows d x d ( x 2 e x ) = n ≥ 0 ∑ d x d ( n ! x n − 2 ) = n ≥ 0 ∑ n ! ( n − 2 ) x n − 3 . After that, we multiply by x at both sides of the equation and then, take another derivative, at both sides, in respect to x (we can still interchange derivative with sum, because the sum is still absolutely convergent), d x d ( x d x d ( x 2 e x ) ) = n ≥ 0 ∑ d x d ( n ! ( n − 2 ) x n − 2 ) = n ≥ 0 ∑ n ! ( n − 2 ) 2 x n − 3 . Then, we can make x = 1 and rewrite the right-hand side of the above equation as 4 + 1 + n ≥ 1 ∑ ( n + 2 ) ! n 2 , and solve the left-hand side of the equation, which give us [ d x d ( x d x d ( x 2 e x ) ) ] x = 1 = 2 e . Finally, we can conclude that our desired sum n ≥ 1 ∑ ( n + 2 ) ! n 2 = 2 e − 5 .
Problem Loading...
Note Loading...
Set Loading...
The given series converges absolutely and the convergence can be verified using ratio test. The given sum can be written as,
S = 3 ! 1 2 + 4 ! 2 2 + 5 ! 3 2 + … = k = 3 ∑ ∞ k ! ( k − 2 ) 2 ⟹ S = k = 3 ∑ ∞ k ! k 2 − 4 k + 4 ⟹ S = k = 3 ∑ ∞ k ! k 2 + k = 3 ∑ ∞ ( − k ! 4 k + k ! 4 ) ⟹ S = k = 3 ∑ ∞ ( k − 1 ) ! k + k = 3 ∑ ∞ ( − ( k − 1 ) ! 4 + k ! 4 ) ⟹ S = k = 3 ∑ ∞ ( k − 1 ) ! ( k − 1 ) + 1 + k = 3 ∑ ∞ ( − ( k − 1 ) ! 4 + k ! 4 ) ⟹ S = k = 3 ∑ ∞ ( k − 2 ) ! 1 + k = 3 ∑ ∞ ( k − 1 ) ! 1 + k = 3 ∑ ∞ ( − ( k − 1 ) ! 4 + k ! 4 ) ⟹ S = k = 1 ∑ ∞ k ! 1 + k = 2 ∑ ∞ k ! 1 + k = 3 ∑ ∞ ( − ( k − 1 ) ! 4 + k ! 4 )
Now, recall the Maclaurin expansion of e x which converges ∀ x . For the case of x = 1 , we get the series for e which is,
Also, note that the last sum (the one on the extreme right in S ) is a telescoping sum and as such, the sum S can be evaluated as,
S = ( e − 1 ) + ( e − 2 ) + ( − ( 3 − 1 ) ! 4 ) = e − 1 + e − 2 − 2 = 2 e − 5
S = 2 e − 5 ≈ 0 . 4 3 6 5 6 4 ⟹ 1 0 0 0 S ≈ 4 3 6 . 5 6 4 ⟹ ⌊ 1 0 0 0 S ⌋ = 4 3 6
Note: The "sum-breaking" done in 3rd and 6th lines of the calculation is valid because the sums obtained after "breaking" are also convergent sums (can be easily verified using convergence tests).