Exponential Series Convergence

Calculus Level 2

If x 1 , x 2 , x 3 , x_1,x_2,x_3,\ldots is a sequence of positive reals such that k = 1 x k \displaystyle \sum_{k=1}^\infty x_k converges, does k = 1 ( e x k 1 ) \sum_{k=1}^{\infty}\left(e^{x_k}-1\right) converge?

Yes, always No, never It depends on the x k x_k

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.

2 solutions

Eli Ross Staff
Nov 13, 2016

Note that e x 1 = x + x 2 2 ! + x 3 3 ! + . e^{x} - 1 = x+ \frac{x^2}{2!} + \frac{x^3}{3!} + \cdots. Since k = 1 x k \sum_{k=1}^\infty x_k converges, lim k x k = 0. \lim_{k\rightarrow\infty}x_k = 0. Thus, for sufficiently large k K , k\ge K, 0 < x k < 1 , 0< x_k < 1, implying that x k > x k 2 > x k 3 > . x_k>x_k^2 > x_k^3 > \cdots. So, for k K , k \ge K, e x k 1 = x k + x k 2 2 ! + x k 3 3 ! + < x k + x k 2 ! + x k 3 ! + = x k ( 1 + 1 2 ! + 1 3 ! + ) = x k ( e 1 ) . e^{x_k} -1 = x_k+ \frac{x_k^2}{2!} + \frac{x_k^3}{3!} + \cdots < x_k+ \frac{x_k}{2!} + \frac{x_k}{3!} + \cdots = x_k\left(1 + \frac{1}{2!} + \frac{1}{3!} + \cdots\right) = x_k(e-1).

The sum k = 1 K 1 ( e x k 1 ) \sum_{k=1}^{K-1} \left(e^{x_k}-1\right) is finitely many finite terms; then, k = K ( e x k 1 ) < ( e 1 ) k = K x k , \sum_{k=K}^{\infty} \left(e^{x_k}-1\right) < (e-1)\cdot \sum_{k=K}^{\infty} x_k, and the right hand side converges since k = 1 x k \sum_{k=1}^\infty x_k converges.

Since we can choose K K such that, 0 < x k < 1 , k K 0<x_k<1, \forall k \geq K , we have e x k 1 + 2 x k , k K . e^{x_k}\leq 1+2x_k, \forall k\geq K. This follows because, for all 0 y 1 0 \leq y \leq 1 , we have the inequality exp ( y ) 1 + 2 y \exp(y) \leq 1+2y , which can be verified using calculus. Thus, k K ( exp ( x k ) 1 ) 2 k K x k < \sum_{k\geq K}(\exp(x_k)-1)\leq 2\sum_{k\geq K} x_k <\infty

Abhishek Sinha - 4 years, 6 months ago

Log in to reply

I think some people won't immediately see why e x k 1 + 2 x k e^{x_k} \le 1 + 2x_k for sufficiently large k . k. Would you mind explaining with slightly more detail?

Eli Ross Staff - 4 years, 6 months ago

Log in to reply

Thanks. I have expanded my comment.

Abhishek Sinha - 4 years, 5 months ago

Ah, very nice comparison!

Calvin Lin Staff - 4 years, 7 months ago

Log in to reply

The motivation comes from using the approximation e x = 1 + x e^x = 1+x for small x x ... "Well, how much is e x e^x like 1 + x ? 1+x? "

The question could probably be generalized with Taylor series of arbitrary functions.

Eli Ross Staff - 4 years, 7 months ago
Abdelhamid Saadi
Nov 23, 2016

let consider f ( y ) = e y f(y) = e^y over the interval [ 0 , x ] [0, x] for x x positive.

By the Mean value theorem , it exists c in [ 0 , x ] [0, x] such: f ( c ) = f ( x ) f ( 0 ) x 0 f'(c) = \dfrac {f(x) - f(0)}{x - 0} This yields: e x 1 x = e c < e x \dfrac {e^x - 1}{x}\ = e^c < e^x

So that for x x in [ 0 , 1 ] [0,1] e x 1 < e × x {e^x - 1} < e \times x

Since k = 1 x k \sum_{k=1}^\infty x_k converges, lim k x k = 0. \lim_{k\rightarrow\infty}x_k = 0.

Thus, for sufficiently large k K , k\ge K, 0 < x k < 1 , 0< x_k < 1, implying that for k K , k \ge K, e x k 1 < x k × e . e^{x_k} -1 < x_k \times e.

by the Comparison Test of convergence k = 1 ( e x k 1 ) \sum_{k=1}^{\infty}\left(e^{x_k}-1\right) est convergente.

How does it come e^c < e^x? Also
e^x -1< e*x???

Saeed Khan - 4 years, 6 months ago

Log in to reply

c ( 0 , x ) c < x e c < e x c \in {(0, x)} \implies c < x \implies e^c < e^x

e x 1 x < e x e x 1 < e x \dfrac {e^x - 1}{x}\ < e^x \implies e^x - 1 < e*x

Abdelhamid Saadi - 4 years, 6 months ago

Can some give explanation?

Saeed Khan - 4 years, 6 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...