Am 07.12.2016 um 14:42 schrieb Allin Cottrell:
On Wed, 7 Dec 2016, Sven Schreiber wrote:
> and/or functions, this caused some complications. So I wanted to
ask
> if it's also possible to run such tests on vectors/matrices, instead
> of series in workfiles. Besides "normtest" another candidate might be
> "difftest", and probably some others, too.
Yes, good idea. To allow such flexibility I think we'd want to add a
function version of normtest (and possibly some other commands). A
normtest() function could accept a series (either named or "on the fly")
or vector, plus an integer to select the test-type, and could return a
2-vector of test statistic and p-value. This should be pretty easy since
it would use the same backend as the command.
Sounds good -- I'm wondering whether it might be a good idea to have a
string argument to select the test type instead of an integer. What I
mean is that 'normtest ... --swilk' would translate directly into
'normtest(..., "swilk")'. (But I can also see some advantages of the
integer-based interface.)
thanks,
sven