Binomial Data
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Binomial Data



Dear members of the ASREML discussion group:


Similar to Hugo Montaldo I am analyising binomial trait (0 and 1's) and I
believe it has a direct and maternal (genetic and permanent) effects. I
obtained the following results with the !logit option:
 
Variance heterogenity factor [Deviance/DF]     0.68

   Source               Model  terms     Gamma     Component    Compnt/StndErr
  ide(dam)              22236   5278  0.289821      0.289821          3.35 U  
  anim                US=UnStr     1  0.937505E-01  0.937505E-01      2.32 U  
  anim                US=UnStr     1 -0.251243E-01 -0.251243E-01     -0.51 U  
  anim                US=UnStr     2  0.129291      0.129291          1.60 U  
 Covariance/Variance/Correlation Matrix US=UnStructu
 0.9375E-01-0.2282    
-0.2512E-01 0.1293    

I understand that ide(dam)=0.289821 is the variance of maternal permanent
effect, anim=0.937505E-01 is the variance of direct effect,
anim=-0.251243E-01 is the covariance between direct and maternal genetic
effect and anim=0.129291 is the variance of maternal genetic effect.

How can I estimate heritabilities for direct and maternal genetic effects
and their s.e.'s in both the original and underlying scale?.

Thanks  

Nicolas