PROC X12 for seasonal adjustment

OceanDream — Tue, 19 Apr 2016 02:27:17 GMT

I'm trying to use proc x12 to adjust seasonality. I use the following code. My data is quarterly.

proc x12 data=HAVE date=yyqq interval=4 noprint;
by company_id;
var assets;
x11 mode=add outforecast;
output out=WANT;
run;

There is error showing "Time ID variable DATADATE contains gaps." This is because some of my observations ended the third quarter of the year instead of the fourth quarter of the year. Is there any way to fix this??

Also, those firms with adjusted assets results have the exact same values of adjusted assets with unadjusted assets. Is this because none of the firm in my data has seasonality pattern in terms of assets OR there is some mistake in my coding? Please help.

Re: PROC X12 for seasonal adjustment

mitrov — Wed, 27 Apr 2016 20:41:59 GMT

You can use PROC TIMESERIES to covert your data into a proper time series with no gaps in the dates.

For example, this will set to 0 the values of all numerical variables for the dates for which there are no observations. You can look at the SETMISS= option in the manual to see other options.

 proc timeseries data=have out=mseries;
      by company_id;
      id yyqq interval=quarter
                   accumulate=total
                   setmiss=0
      var _numeric_;
  run;

topic Re: PROC X12 for seasonal adjustment in SAS Forecasting and Econometrics

PROC X12 for seasonal adjustment

Re: PROC X12 for seasonal adjustment