Message Boards Message Boards

0
|
7230 Views
|
4 Replies
|
0 Total Likes
View groups...
Share
Share this post:

[?] Why does this series have imaginary terms?

Posted 7 years ago

I am trying to construct a series around infinity and I get the following result:

In[14]:= Normal@Series[ Sqrt[2 Pi x] BesselI[3, x]/Exp[x], {x, \[Infinity], 3}, Assumptions -> (x > 0)]
Out[15]=(-I + E^(2*x)*(1 + 945/(128*x^2) - 35/(8*x)) - (945*I)/(128*x^2) - (35*I)/(8*x))/E^(2*x)

Notice that x is assumed real which means that the result should not not have any imaginary terms but it does. What is going on here? And what is the best way to get rid of these terms?

POSTED BY: David Sagan
4 Replies

The asymptotic expansions for modified Bessel functions are inherently complex functions. See asymptotic expansion for I_nu(z) in http://functions.wolfram.com/Bessel-TypeFunctions/BesselI/introductions/Bessels/ShowAll.html

?? ?

POSTED BY: S M Blinder
Posted 7 years ago

Agreed. But the series is done with the assumption that x is real and under this assumption the imaginary part is zero. But what Mathematica shows is a non-zero imaginary part. This should not be.

POSTED BY: David Sagan

Try:

ComplexExpand@Re@Normal@Series[Sqrt[2 Pi x] BesselI[3, x]/Exp[x], {x, \[Infinity], 3}, Assumptions -> {x > 0}]
POSTED BY: Mariusz Iwaniuk
Posted 7 years ago

OK that gets rid of the imaginary terms.

Much Thanks!

Anyone have any ideas as to why there are imaginary terms to begin with?

POSTED BY: David Sagan
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract