Date:         Fri, 17 Jan 1997 16:43:37 -0800
Reply-To:     Joe Lin <joelin@QNTM.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Joe Lin <joelin@QNTM.COM>
Subject:      Method, procedure to use when empty cells in  ANOVA?
Content-Type: text/plain; charset=US-ASCII

     Hi,
 
     I have a multi-factor data with empty cells. For example, I have three
     levels for factor A and 4 levels for factor B. Now I have a 3 by 4
     matrix. But I don't have observation in couple of the combinations of
     A and B factors. The matrix is incomplete and has empty cells. I would
     still want to fit a linear model and get the least square estimates
     and the interval estimates for the factor level means. Any suggestion
     of what method I can use and the procedure that is available in SAS?
 
     Thanks.
 
     Joe