Binary Data with missing values

Dear All

I wonder if someone could provide advice on dealing with missing values
in binary data models.

I am in the process of performing association studies between SNPs and
pregnancy status in cows. I have genotypic information on approximately
90% of animals and pregnancy status data on approximately 80% of animals
(coded as 0 = pregnant, 1 = not pregnant).

At the moment, my model is very simple -

Fertility Analysis

 ANIMALID !P

 Sire !P

 Dam !P

 Herd !I

 Age

 PregnancyStatus

 SNP1!I

Pedigree.ped !MAKE !SORT !REPEAT

Pregnancy.csv !CSV !MAXIT 100 !EXTRA 20 !MVINCLUDE !SKIP 1 !DISPLAY 15

PregnancyStatus !BIN ~ mu Herd Age SNP1 !r ANIMALID

The model only runs when the !MVINCLUDE qualifier is specified, however,
I am trying to find a way of excluding missing values, as including them
interferes with the result. Using MVEXCLUDE / MVREMOVE causes the model
to fail.

I would be very grateful for any replies,

Many thanks

Andrew

 
Received on Thu Jul 08 2009 - 11:16:07 EST