fitCARBayesST

a function to model curves using a spatio-temporal bayesian modelling using CARBayesST library

fitCARBayesST(
  datain,
  xvar,
  trait,
  k = NULL,
  graphDist,
  burnin = 500,
  n.sample = 1500,
  formulaModel,
  typeModel = "anova",
  verbose
)

Arguments

datain	input dataframe of parameters
xvar	character, time variable (ex: thermalTime)
trait	character, parameter of interest (ex: plantHeight)
k	numeric, number of nearest neighbours to be returned
graphDist	logical, display distance graph
burnin	The number of MCMC samples to discard as the burn-in period
n.sample	The number of MCMC samples to generate
formulaModel	A formula for the covariate part of the model using the syntax of the lm() function. see ST.CARanova help
typeModel	character, choice of the model, anova, linear or ar
verbose	logical FALSE by default, if TRUE display information about the progress

Value

a list with a spatio-temporal object (CARBayesST) and a dataframe of the formated data

Details

the input dataset must contain Position,Line,Ref,scenario,genotypeAlias columns. The function is not generic and needs specific columns names in the input data set. Please have a look of the struture of the data set used in the example.

See also

CARBayesST, ST.CARanova and spdep

Examples

 data(plant1)
 library(dplyr)
#> 
#> Attachement du package : 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
 mydata<-filter(plant1,!is.na(plant1$thermalTime))
 str(mydata)
#> 'data.frame':	45342 obs. of  14 variables:
#>  $ Ref            : Factor w/ 1680 levels "manip1_10_10_WW",..: 131 131 131 131 131 131 131 131 131 131 ...
#>  $ experimentAlias: Factor w/ 1 level "manip1": 1 1 1 1 1 1 1 1 1 1 ...
#>  $ Day            : Factor w/ 42 levels "2013-02-01","2013-02-02",..: 3 4 5 6 7 9 9 10 11 12 ...
#>  $ potAlias       : int  1 1 1 1 1 1 1 1 1 1 ...
#>  $ scenario       : Factor w/ 2 levels "WD","WW": 2 2 2 2 2 2 2 2 2 2 ...
#>  $ genotypeAlias  : Factor w/ 274 levels "11430_H","A310_H",..: 165 165 165 165 165 165 165 165 165 165 ...
#>  $ repetition     : int  1 1 1 1 1 1 1 1 1 1 ...
#>  $ Line           : int  1 1 1 1 1 1 1 1 1 1 ...
#>  $ Position       : int  1 1 1 1 1 1 1 1 1 1 ...
#>  $ thermalTime    : num  1.29 2.65 3.98 5.32 6.66 ...
#>  $ plantHeight    : num  140 151 213 239 271 ...
#>  $ leafArea       : num  0.018 0.019 0.0208 0.0222 0.0235 ...
#>  $ biovolume      : num  0.253 0.62 1.201 1.68 3.396 ...
#>  $ Repsce         : Factor w/ 15 levels "1_WD","1_WW",..: 2 2 2 2 2 2 2 2 2 2 ...
 test<-fitCARBayesST(datain=mydata,xvar="thermalTime",
       trait="plantHeight",k=2,
       graphDist=TRUE,burnin=10,n.sample=110,
       formulaModel=as.formula(plantHeight~scenario+genotypeAlias),
       typeModel="anova",verbose=FALSE)
#> [1] 22.35558
#>            used  (Mb) gc trigger   (Mb) max used  (Mb)
#> Ncells  3713681 198.4    6893817  368.2  3713681 198.4
#> Vcells 67183832 512.6  211775864 1615.8 67183832 512.6