Post by sahir on Aug 31, 2017 15:14:29 GMT
As mentioned in this FAQ Can I run a GBLUP prediction analysis with GCTA? it seems like its possible to calculate the residuals using the code below.
Since no covariates are included, I am assuming the last column of the output (which is said to be the residual) is calculated as
residual = y - intercept - g - e
is this correct?
I am asking for this clarification because the model in that thread doesn't include the intercept (y = g + e). However when running the code above, I notice the output says:
Is the 1 fixed effect(s) referring to the intercept?
I am trying to use the residuals for two-step methods, where the 2nd step involves fitting a multivariate penalized regression on the residuals calculated in step 1 (after having removed the family effect)
gcta64 --reml --grm test --pheno test.phen --reml-pred-rand --out test Since no covariates are included, I am assuming the last column of the output (which is said to be the residual) is calculated as
residual = y - intercept - g - e
is this correct?
I am asking for this clarification because the model in that thread doesn't include the intercept (y = g + e). However when running the code above, I notice the output says:
Performing REML analysis ... (Note: may take hours depending on sample size).
1085 observations, 1 fixed effect(s), and 2 variance component(s)(including residual variance).
Calculating prior values of variance components by EM-REML ...
1085 observations, 1 fixed effect(s), and 2 variance component(s)(including residual variance).
Calculating prior values of variance components by EM-REML ...
Is the 1 fixed effect(s) referring to the intercept?
I am trying to use the residuals for two-step methods, where the 2nd step involves fitting a multivariate penalized regression on the residuals calculated in step 1 (after having removed the family effect)