| To: | s-news@lists.biostat.wustl.edu |
|---|---|
| Subject: | NA's in LME |
| From: | Pravin Jadhav <pravinj@gmail.com> |
| Date: | Wed, 14 Dec 2005 11:25:56 -0500 |
| Domainkey-signature: | a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:mime-version:content-type; b=MrxHWnWbfCQfGlU7jWpwdLTz7rgecjg0+Ie1rmdPk+Vs2aefspG4ETVFOP7mXC9lhBH3HR8AHknA7dn0rRwjkimfIMzObcLOFEOzuHv/YaIA0zMc5MOA+9zmWbWVpfT9UumSeTrtcxSJ8OUCfQzO5upSRERMv61eCBaHc6wJjI0= |
|
Hello, In LME, the default action for the missing data is 'na.fail'. My aim is to obtain predictions for the missing data based on individual random effects. One option would be to get the output and do the calculations based on random coefficients. I was wondering if there is any direct way (like NONMEM) in Splus. I tried 'na.include' but did not work. Is this possible in Splus? Any suggestions? Thanks, Pravin |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: call c++ from splus, Horace Tso |
|---|---|
| Next by Date: | How to stack, Overstreet, Jason (FL51) |
| Previous by Thread: | call c++ from splus, Tony Smith |
| Next by Thread: | How to stack, Overstreet, Jason (FL51) |
| Indexes: | [Date] [Thread] [Top] [All Lists] |