I tried it in Stata. the command display (-.3333)^(1/3) produces an
error message. Removing the second set of parentheses yields -.3333.
>
> Thanks. I did not know this paragraph existed. However, all it did was
> tell me that SPSS cannot carry out the compute, which I already knew,
> but not why. All the domain errors they list, I believe, except
> (-1)**(1/3) and a number too large to be represented in the computer,
> are not mathematically defined and so I would not expect SPSS to be able
> to carry out a calculation. However, (-1)**(1/3) is mathematically
> defined. I tried this on my HP 35s hand calculator and it came up with
> an error. This did not seem unreasonable because this is an inexpensive
> calculating machine. I then went to SPSS, a more advanced software, and
> discovered SPSS cannot do it either. Does anyone know why? Is it not
> possible with the language upon which SPSS is based? I wonder if SAS can
> do it. Does anyone know?
> Stan
>
> On 11/15/2019 2:12 AM, Kirill Orlov wrote:
> > See DOMAIN ERRORS paragraph in Command Syntax Reference.
> >
> >
> > 15.11.2019 4:53, Stan Gorodenski пишет:
> >> compute z = (-1)**(1/3).
> >
> >
> >
> >
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
>
[hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
command. To leave the list, send the command