Dear All,

I have  looked at this problem with  Zelig previously, but still  cannot  find how to  combine the  five  datasets  produced by AmeliaView.   In our  data it is critical to set bounds for factors, otherwise the imputed data produces out of range values for standard blood tests. How exactly  do you  create the output object that is needed to combine the five  datasets that are  produced by AmeliaView? 

Although page  27 of the Amelia user's manual  gives the  instructions for setting bounds without using the AmeliaView GUI, the  syntax seems cumbersome "  bds <- matrix(c(3, 30, 40), nrow = 1, ncol = 3) " if this must be  done for each variable. The alternative  presented, to allow unbounded  imputation  that might produce  negative  values,  that would then be set to  "0" would  produce unacceptable  results.... such  patients  would be  dead.

I  very much  appreciate your efforts with  this program.  I and my colleagues at the Wistar Institute of the University  of Pennsylvania have  successfully used the older version  of  Amelia and Zelig  for a manuscript on  pediatric treatment response for HIV that has been accepted pending revisions by  PLoS- One, so I hope to be able to  continue to use the program. I  even tried to load the older, more user friendly version  of Zelig , but it would not work with the latest version  of R. Any additional guidance as to how the bounded datasets produced by AmeliaView  can  be turned into an output  object and then combined would be much  appreciated. Perhaps a new updated version of this excellent package (named "Hoffa" ?) might simplify this process.

Skip  Barbour