st: Tssmooth Ma Missing Values

[<J.M.Chwieroth@lse.ac.uk>]

I'm trying to generate 2 and 5 year forward and backward moving averages for a particular variable.   A sample of the data structure appears below.  To generate the moving averages below I have used the following syntax:
 
tssmooth ma 2yearMA=Variable, window(0 1 1)

tssmooth ma 5yearMA=Variable, window(0 1 4)

tssmooth ma Back2YearMa=Variable, window(1 1 0)

 

The problem I encounter is that the forward moving average syntax leads Stata to create new values prior to the start of the times in rows where I am missing data for the variable.  At the end of the time series the same occurs for the backward moving average.  Ideally, I want Stata to produce moving averages only in years for which I have data for the variable.  

Is there some syntax I could use to tell Stata not to create these new values in rows where I have missing data? 

 
 
Country	 Year	 Variable	 2yearMA	 5YearMa	 Back2YearMa	
Australia	 1897	 .	 .	 0	 .	 	
Australia	 1898	 .	 .	 0.5	 .	 	
Australia	 1899	 .	 .	 1	 .	 	
Australia	 1900	 .	 0	 1.5	 .	 	
Australia	 1901	 0	 0.5	 2	 .	 	
Australia	 1902	 1	 1.5	 3	 0	 	
Australia	 1903	 2	 2.5	 4	 0.5	 	
Australia	 1904	 3	 3.5	 5	 1.5	 	
Australia	 1905	 4	 4.5	 6	 2.5	 	
Australia	 1906	 5	 5.5	 7	 3.5	 	
Australia	 1907	 6	 6.5	 8	 4.5	 	
Australia	 1908	 7	 7.5	 9	 5.5	 	
Australia	 1909	 8	 8.5	 10	 6.5	 	
Australia	 1910	 9	 9.5	 11	 7.5	 	
Australia	 1911	 10	 10.5	 12	 8.5	 	
Australia	 1912	 .	 .	 .	 9.5	 	

Please access the attached hyperlink for an important electronic communications disclaimer: http://www.lse.ac.uk/collections/planningAndCorporatePolicy/legalandComplianceTeam/legal/disclaimer.htm

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/