Compositional inverse of multivariate power series

28 views
Skip to first unread message

Peter Luschny

unread,
Feb 10, 2021, 6:44:09 AM2/10/21
to sage-support
The compositional inverse (with respect to x) of 

    y(t, x) = x - t*(exp(x) - 1)

is

    1/(1-t)*y + t/(1-t)^3*y^2/2! + (t+2*t^2)/(1-t)^5*y^3/3! + (t+8*t^2+6*t^3)/(1-t)^7*y^4/4! + ...

Apparently multivariate power series rings do not know how 
to reverse a series.     

Perhaps there is a workaround?
Thanks!


Dima Pasechnik

unread,
Feb 10, 2021, 7:53:38 AM2/10/21
to sage-support
one can try to see if working with univariate series over R(t) will
give the needed inverse.


> Thanks!
>
>
> --
> You received this message because you are subscribed to the Google Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sage-support...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sage-support/b70394ce-3b11-412e-8a1d-3acd89ace6c6n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages