summary.SpATS: Summary method for 'SpATS' objects

In SpATS: Spatial Analysis of Field Trials with Splines

Summary method for SpATS objects

Description

Default summary method for objects fitted with SpATS() function.

Usage

## S3 method for class 'SpATS'
summary(object, which = c("dimensions", "variances", "all"), ...)

Arguments

object	an object of class SpATS as produced by SpATS()
which	character vector indicating the information to be shown on the screen. “dimensions”: prints the dimensions associated to each component of the fitted model (see details), “variances”: prints the variance components and standard deviation (SD), the residual variance, and the ratio between the residual variance and the variance components (in log10 scale); and, “all”: both dimensions and variances are printed.
...	further arguments passed to or from other methods. Not yet implemented.

Details

This function provides relevant information about the fitted model. The dimensions associated to each model term, as well as the variance components (jointly with the standard deviation), can be obtained. Specifically, the following information is shown:

dimensions

dimensions associated to each model term (genotype, spatial, fixed and/or random):

• Effective: the effective dimension or effective degrees of freedom. The concept of effective dimension is well known in the smoothing framework. However, its use has been less recognized in the mixed model framework. The effective dimension of a fitted model is defined as the trace of the so-called hat matrix (Hastie and Tibshirani, 1990). For each model term, the effective dimension is then defined as the trace of the block of the hat matrix corresponding to this term. For fixed effects, the effective dimension coincides with the model dimension (see below). For random and smooth terms, however, the effective dimension is usually lower than the model dimension. In these cases, the effective dimension or effective degrees of freedom can be interpreted as the effective number of estimated parameters. It is worth nothing that for the specification of the spatial trend using the SAP function, the effective dimension associated to each spatial coordinates is shown (see Rodriguez-Alvarez et al. (2015) for details), as well as the global effective dimension (the sum of these two values). For the ANOVA-type decomposition using SAP, besides the former information, the effective dimension associated to each main effect is also shown. This is in concordance with the number of smoothing parameters (or variance components in the mixed model framework), used to model the spatial trend. Accordingly, for the spatial trend specified using the function PSANOVA, the effective dimension of each of the five components is shown (and no global summary is reported). More information can be found in Rodriguez-Alvarez et al. (2018).

• Model: the model dimension corresponds to the number of parameters to be estimated. For the specification of the spatial trend using the SpATS function, the model and nominal dimensions are only specified for the global term (and for the main effects when using the ANOVA-type decomposition).

• Nominal: for the random terms of the model, the nominal dimension corresponds to upper bound for the effective dimension (i.e., the maximum effective dimension a random term can achive). This nominal dimension is rank[X, Z_k] - rank[X], where Z_k is the design matrix of the k-th random factor and X is the design matrix of the fixed part of the model. In most cases (but not always), the nominal dimension corresponds to the model dimension minus one, “lost” due to the implicit constraint that ensures the mean of the random effects to be zero.

• Ratio: ratio between the effective dimension and the nominal dimension. For the genotype (when random), this ratio corresponds to the generalized heritability proposed by Oakey (2006). A deeper discussion can be found in Rodriguez - Alvarez et al. (2018)

• Type: Model term type 'F' Fixed, 'R' Random, and 'S' Smooth (spatial trend).

variances

variance components associated to the random terms and the smooth spatial trend, as well as the ratio between the residual variance and the variance components (in log10 scale). In the smoothing framework, this ratio corresponds to the smoothing parameter.

Value

Returns an object of class summary.SpATS with the same components as an SpATS object (see SpATS) plus:

p.table.dim	a matrix containing all the information related to the dimensions of the fitted model.
p.table.vc	a matrix containing all the information related to the variance components of the fitted model

References

Hastie, T., and Tibshirani, R. (1990). Generalized Additive Models. In: Monographs on Statistics and Applied Probability, Chapman and Hall, London.

Oakey, H., A. Verbyla, W. Pitchford, B. Cullis, and H. Kuchel (2006). Joint modeling of additive and non-additive genetic line effects in single field trials. Theoretical and Applied Genetics, 113, 809 - 819.

Rodriguez-Alvarez, M.X, Boer, M.P., van Eeuwijk, F.A., and Eilers, P.H.C. (2018). Correcting for spatial heterogeneity in plant breeding experiments with P-splines. Spatial Statistics, 23, 52 - 71. https://doi.org/10.1016/j.spasta.2017.10.003.

Rodriguez-Alvarez, M.X., Lee, D.-J., Kneib, T., Durban, M., and Eilers, P.H.C. (2015). Fast smoothing parameter separation in multidimensional generalized P-splines: the SAP algorithm. Statistics and Computing, 25, 941 - 957.

See Also

SpATS

Examples

library(SpATS)
data(wheatdata)
summary(wheatdata)

# Create factor variable for row and columns
wheatdata$R <- as.factor(wheatdata$row)
wheatdata$C <- as.factor(wheatdata$col)

m0 <- SpATS(response = "yield", spatial = ~ SAP(col, row, nseg = c(10,20), degree = 3, pord = 2), 
 genotype = "geno", fixed = ~ colcode + rowcode, random = ~ R + C, data = wheatdata, 
 control =  list(tolerance = 1e-03))
# Brief summary
m0
# More information: dimensions
summary(m0) # summary(fit.m2, which = "dimensions") 
# More information: variances
summary(m0, which = "variances") 
# More information: all
summary(m0, which = "all") 

SpATS documentation built on Nov. 10, 2022, 5:58 p.m.