Re: NAs in the response variable

Dear saidou

I fitted a mixed linear model with ASREML-R, and I have missing data in
both response variable (Y) and in one factor (X). For data in only one
environment (none spatial data) I used the options na.method.Y="omit" to
exclude observations with NAs.
na.method.Y="omit" did not work out, and give exacty the same fit than
na.method.Y="include". So I tried na.method.Y="exclude" which gave a
different result, and seemed to work well. The option given in ASREML-R
pdf for exluding NAs in Y is "omit" (and not "exclude"). Was the term
"omit" written erroneously in the pdf manual instead of "exclude" perhaps?

If not, what can lead to the situation "omit" option worked exactly as
"include" in the presence of NAs in Y?

$$$$$
There is a detailed description of the above on p71 of the asreml manual
(you can access this in an R session by typing

> asreml.man()

which opens up the manual using pdf() as a default.

The manual states that you can use "omit", "fail" or "include" for both
na.method.Y and na.method.X. The defaults for these are "include" and
"fail" respectively. Hence this is why your job stopped, as you need to
specify both with "include" if you wish to include all data with missing
values in y and X

Hope this helps (HTH)

warm regards

Brian Cullis|Research Leader, Biometrics &
Senior Principal Research Scientist |
Industry & Investment NSW | Wagga Wagga Agricultural Institute | Pine
Gully Road | Wagga Wagga NSW 2650 |
PMB | Wagga Wagga NSW 2650
T: 02 6938 1855 | M: 0439 448 591 | F: 02 6938 1809 | E:
brian.cullis_at_industry.nsw.gov,au
W: www.industry.nsw.gov.au |

Visiting Professorial Fellow
School of Mathematics and Applied Statistics
Faculty of Informatics
University of Wollongong

Professor,
Faculty of Agriculture, Food & Natural Resources
The University of Sydney

Adjunct Professor
School of Computing and Mathematics
Charles Sturt University
 
Received on Sat Oct 16 2009 - 09:33:47 EST