asremlPlus (version 2.0-2)

asremlPlus-package: Augments the use of asreml-R in fitting mixed models

Description

Assists in automating the testing of terms in mixed models when 'asreml' is used to fit the models. The content falls into the following natural groupings: (i) Data, (ii) Object manipulation functions, (iii) Model modification functions, (iv) Model testing functions, (v) Model diagnostics functions, (vi) Prediction production and presentation functions, (vii) Response transformation functions, and (viii) Miscellaneous functions. A history of the fitting of a sequence of models is kept in a data frame. Procedures are available for choosing models that conform to the hierarchy or marginality principle and for displaying predictions for significant terms in tables and graphs. The package 'asreml' provides a computationally efficient algorithm for fitting mixed models using Residual Maximum Likelihood. It can be purchased from 'VSNi' (http://www.vsni.co.uk/) as 'asreml-R', who will supply a zip file for local installation/updating.

Details

The DESCRIPTION file: asremlPlus Index: (i) Data Wheat.dat Data for an experiment to investigate 25 varieties of wheat. (ii) Object manipulation alldiffs Forms an object of S3-class 'alldiffs' that stores the predictions for a model fitted using asreml, along with statistics for all pairwise differences. asrtests Forms an object of S3-class 'asrtests' that stores a fitted asreml object, a pseudo-anova table for the fixed and a history of changes and hypthesis testing used in obtaining the model. print.alldiffs Prints the values in an 'alldiffs' object in a nice format. print.asrtests Prints the values in an 'asrtests' object (iii) Model modification addrm.terms.asrtests Adds or removes the specified set terms from either the fixed or random model and records the change in a data.frame. newfit.asreml Refits an asreml model with modified model formula using either a call to 'update.asreml' or a direct call to 'asreml'. newrcov.asrtests Fits a new rcov formula using asreml. rmboundary.asrtests Removes any boundary or singular variance components from the fit stored in 'asreml.obj' and records their removal in a data.frame. setvarianceterms.asreml Allows the seting of constraints and initial values for terms in the 'random' and 'rcov' arguments of an 'asreml' call. sig.devn.reparam.asrtests This function reparamterizes each random (deviations) term involving 'devn.fac' to a fixed term and ensures that the same term, with 'trend.num' replacing 'devn.fac', is included if any other term with 'trend.num' is included in 'terms'. (iv) Model testing choose.model.asrtests Determines the set of significant terms taking into account hierarchy or marginality relations. info.crit.asreml Computes AIC and BIC for a model. recalc.wald.tab.asrtests Recalculates the denDF, F.inc and P values for a table of Wald test statistics obtained using 'wald.asreml'. reml.lrt.asreml Performs REML likelihood ratio test. testranfix.asrtests Tests for a single fixed or random term in model fitted using 'asreml'. testrcov.asrtests Fits a new rcov formula using 'asreml' and tests whether the change is significant. testswapran.asrtests Tests, using a REMLRT, whether the difference between current random model and one in which oldterms are dropped and newterms are added is significant. (v) Model diagnostics variofaces.asreml Plot empirical variogram faces, including envelopes, as described by Stefanova, Smith & Cullis (2009) (vi) Prediction production and presentation pred.present.asreml This function forms the predictions for each significant term and presents them in tables and/or graphs. predictiondiffs.asreml Forms all pairwise differences between a set of predictions, their standard errors and p-values for a test of whether the differences are significantly different from zero. predictionplot.asreml This function plots the predictions for a term, possibly with error bars. predictparallel.asreml This function forms the predictions and associated statistics for a term, taking into account that a numeric vector and a factor having parallel values may occur in the model. It stores the results in an object of class 'alldifffs' and prints the results. It can be used when there are not parallel values. (vii) Response transformation angular Applies the angular transformation to proportions. angular.mod Applies the modified angular transformation to a vector of counts. power.transform Perform linear and power transformations on a variable whose name is given as a character string in 'var.name'. The transformed variable is stored in the 'data.frame data'. (viii) Miscellaneous num.recode Recodes the unique values of a vector using the values in a new vector. permute.square Permutes the rows and columns of a square matrix. permute.to.zero.lowertri Permutes a square matrix until all the lower triangular elements are zero. The functions whose names end in 'asrtests', which are most of the model functions, utilize an asrtests object that stores: (i) the currently fitted model in asreml.obj, (ii) the table of test statistics for the fixed effects in wald.tab, and (iii) a data frame that contains a history of the changes made to the model in test.summary.

References

Butler, D. G., et al. (2010). Analysis of Mixed Models for {S} language environments: {ASReml-R} reference manual. Brisbane, DPI Publications. Stefanova, K. T., Smith, A. B. & Cullis, B. R. (2009) Enhanced diagnostics for the spatial analysis of field trials. Journal of Agricultural, Biological, and Environmental Statistics, 14, 392--410.

See Also

asreml

Examples

## Analyse wheat dat using asreml and asremlPlus
## Set up for analysis
library(dae)
library(asreml)
## use ?Wheat.dat for data set details
data(Wheat.dat)

# Fit initial model
current.asr <- asreml(yield ~ Rep + WithinColPairs + Variety, 
                      random = ~ Row + Column + units,
                      rcov = ~ ar1(Row):ar1(Column), 
                      data=Wheat.dat)
summary(current.asr)

# Load current fit into an asrtests object
current.asrt <- asrtests(current.asr, NULL, NULL)

# Check for and remove any boundary terms
current.asrt <- rmboundary.asrtests(current.asrt)

#Check term for within Column pairs
current.asrt <- testranfix.asrtests("WithinColPairs", current.asrt, drop.fix.ns=TRUE)

# Test nugget term
current.asrt <- testranfix.asrtests("units", current.asrt, positive=TRUE)

# Test Row autocorrelation
current.asrt <- testrcov.asrtests("~ Row:ar1(Column)", current.asrt, 
                                  label="Row autocorrelation", simpler=TRUE)

# Test Col autocorrelation (depends on whether Row autocorrelation retained)
k <- match("Row autocorrelation", current.asrt$test.summary$terms)
p <- current.asrt$test.summary$p
{if (p[k] <= 0.05)
  current.asrt <- testrcov.asrtests("~ ar1(Row):Column", current.asrt, 
                                    label="Col autocorrelation", simpler=TRUE,
                                    update=FALSE)
 else
   current.asrt <- testrcov.asrtests("~ Row:Column", current.asrt, 
                                     label="Col autocorrelation", simpler=TRUE,
                                     update=FALSE)
}
print(current.asrt)

# Get current fitted asreml object
current.asr <- current.asrt$asreml.obj
current.asr <- update(current.asr, aom=TRUE)


# Do residuals-versus-fitted values plot
plot(fitted.values(current.asr),residuals(current.asr))

# Form variance matrix based on estimated variance parameters
s2 <- current.asr$sigma2
gamma.Row <- current.asr$gammas[1]
gamma.unit <- current.asr$gammas[2]
rho.r <- current.asr$gammas[4]
rho.c <- current.asr$gammas[5]
row.ar1 <- mat.ar1(order=10, rho=rho.r)
col.ar1 <- mat.ar1(order=15, rho=rho.c)
V <- fac.vcmat(Wheat.dat$Row, gamma.Row) + 
     gamma.unit * diag(1, nrow=150, ncol=150) + 
     mat.dirprod(row.ar1, col.ar1)
V <- s2*V

#Produce variogram and variogram faces plot (Stefanaova et al, 2009)
plot.asrVariogram(variogram(current.asr))
variofaces.asreml(current.asr, V=V)

#Get Variety predictionsand all pairwise prediction differences and p-values
Var.diffs <- predictparallel.asreml(classify = "Variety", 
                                    asreml.obj=current.asr,
                                    error.intervals="halfLeast",
                                    wald.tab=current.asrt$wald.tab,
                                    tables = "predictions")
print(Var.diffs, which = c("differences", "p.differences"))