13.07.2015 Views

ASReml-S reference manual - VSN International

ASReml-S reference manual - VSN International

ASReml-S reference manual - VSN International

SHOW MORE
SHOW LESS
  • No tags were found...

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

6.2 The predict method 64The predict function call for this example is:predict(obj, classify = c(’A’,’B’), present = list(’A’ = c(f 1 , f 2 , . . .),’B’ = list(c(f 11 , f 12 , . . .),c(f 21 , f 22 , . . .), prwts = c(f 11 , f 12 , . . .))),sed = list(’A’ = TRUE, ’B’ = FALSE))and is illustrated by the tree diagram in Figure 6.2.1. The problem could of course besimplified somewhat by two separate predict() calls for each of ’A’ and ’B’.Fig. 6.1. Predict tree structurepredict✭✭✭✭✭✭✭✭✭✭✭✪✪❤❤❤❤❤❤❤❤❤❤classifypresentsed ❧ ✏ ✏✏✏✏ ✧❡ ✧✧ ◗ ◗’A’ ’B’ A✟✟✟B❍ A B❍❍f 1f 2list1 list2 prwts TRUE FALSE.f 11f 12.f 21f 22.w 1w 2.6.2.2 The prediction processmv, unitsPredictions are formed as an extra process in the final iteration. predict.asreml() parsesthe argument list and calls update.asreml() using the final parameter estimates in therequired asreml object. Additional arguments to asreml may be included in the call topredict.asreml(), such as requesting extra memory, adding spline predict points or controllingthe number of additional iterations, bound by the rules of update.asreml().By default, factors are predicted at each level, simple covariates are predicted at theiroverall mean and covariates used as a basis for splines or orthogonal polynomials arepredicted at their design points. Covariates grouped into a single term using the grp()model function) are treated as covariates.Special model terms mv and units are always ignored.Prediction at particular values of a covariate or particular levels of a factor is achievedby:1. Including the variables in the classify set and specifying any non-default values atwhich predictions are to be made by using the levels argument.2. Specifying the averaging set. The default averaging set is those explanatory variablesinvolved in fixed effect model terms that are not in the classifying set. By defaultvariables that only define random model terms are ignored. The average argumentallows these variables to be added to the default averaging set.3. Determining the linear model terms to use in prediction. The default rule is that allmodel terms based entirely on the classifying and averaging set are used. The useand ignore arguments allow this default set of model terms to be modified by adding

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!