...interval
For the interested reader, some remarks on endpoints are contained in footnotes throughout.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...x=-2
It is not hard to see that the series diverges in each of these cases by other means. For example, when x=2 we are looking at the sum 22#22. Since the terms of this sum do not tend to 0, the sum will diverge. The case of x=-2 is similar.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...determined
The series converges at both 26#26. The method for showing this is through what is called the integral test, a comparison between the sum and an improper integral.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...x=2
It turns out that the series converges at x=2 but that it diverges at x=0.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...above
With a few technical improvements, the arguments presented here really work for any power series.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...applies
To be honest, there is one case for which this discussion is not quite right. When the radius of convergence is infinite, the series converges even faster than this: the power series converges faster than any geometric series.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

John Jones
Wed Nov 6 10:21:32 MST 1996