On Thu, 18 Jan 2018, Sven Schreiber wrote:
Hi,
I observe the following problem (also with today's snapshot). Right now I
don't have the time to try to create a minimal-example script, but I can
follow up on that. And perhaps the verbal explanation already helps.
In a quarterly workfile I have a couple of monthly variables as Midas lists,
say x_m3/m2/m1 and y_m3/m2/m1.
I then create a single regular (quarterly) variable by calculating something
like:
series z = x_m3 - y_m3
Finally I apply a regular (quarterly) lag:
z = z(-1)
In the end --but only after the last step-- the resulting z series is a
constant (non-unit) term.
z = z(-1)
will always give you a constant series, regardless of how z came
into existence. Think about it!
Allin