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. cheers, sven