x12 set to multiplicative results in additive factors

[afryan]

When I run x12 on a series using the dialog screen, and I set the "X11 Method" setting in "Seasonal Adjustment" tab of the "X12 Options" dialog to "Multiplicative", the result is final seasonal factors that are additive rather than multiplicative. Is there a way to force x12 to calculate multiplicative factors?

The data series has about 103 quarterly observations and has pretty consistent seasonality, and no negative values.

Ultimately, I'm trying to use the following line as part of a command file. But at the moment, I get additive factors whether I run in the command line or in the dialogue screen.
usdem.x12(gdir="C:\USERS\ARYAN003\EV_TEMP\x12graphs\", mode=m, save="b1 d10 d11 d12 d13", arima="(0 1 1)(0 1 1)",tf=auto,reg="td easter[7]",f,flead=10)

Incidently, if I run the following in the command line, I can get multiplicative seasonal factors, but I can't seem to get it to forecast.
usdem.x12(gdir="C:\USERS\ARYAN003\EV_TEMP\x12graphs\", mode=m, save="b1 d10 d11 d12 d13",x11reg="td easter[7]", "f",flead=10,p)

I'm making a shift to Eviews 7 from an earlier version. Since my old x11 command line doesn't work, I'm trying to come up with another solution. Critically, I need to have a five quarter forecast of the seasonal factors, because I need to be able to convert back to the non-seasonally adjusted levels once I complete the forecast. I would use the amdl feature and let x12 pick the arima specification, but that doesn't seem to work for me, as I can't seem to get it to create a forecast more than a year if I set amdl=f.

Why am I trying to force multiplicative factors? I guess because I just want to be sure I know what kind of factors are being generated. If I could ensure it was always additive I guess that would work. Since I'm running x12 in a command line that runs across about 10 series once a quarter, I'd like to know that the seasonal factors are always of one type or another. That way I can convert back to non-seasonally adjusted data in the model with a known equation. My short term solution has been to calculate multiplicative seasonal factors as the ratio NSA / SA, and I can then use those factors in my existing Excel-based workbooks.

After running X12 from the dialog, I get output that starts with the following.

Contents of spc file C:\Users\aryan003\AppData\ev_temp\EVX12TMP.spc

Line #
------
1: series{
2: title = "USDEM"
3: start = 1987.1
4: period = 4
5: name = "USDEM"
6: file = "EVX12TMP.DAT"
7: }
8:
9: transform{
10: function=auto
11: }
12:
13: regression{
14: variables=(td easter[8] )
15: }
16:
17: arima{
18: model=(0 1 1)(0 1 1)
19: }
20:
21: x11{
22: sigmalim = (1.5,2.5)
23: print = ( +ftestd8 +residualseasf +x11diag +qstat +specsa +specirr)
24: save = ( D10 D11 D12 D13)
25: savelog = (q,q2,fb1,fd8,msf)
26: }
27: