st: replacing missing values in one variable with values of another variable

[Alison McCarthy <statslearner@live.com>]

Hi,
I have cross-sectional time series dataset of unemployment, occupation (when employed), and health. There are about 10,000 people and 10 years. 

I want to examine whether a persons last observed occupation influences their health when unemployed. At the moment, there are missing values for occupation if a person is unemployed. I would like to replace these missing values with their last observed occupation. 

Any idea how to do this? 

Best wishes
Alison 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/