Resampling for linear models

Source: R/resampling.model.R

This process consists of finding the values of P-value by means of a re-sampling (permutation) process along with the values obtained by variance analysis.

resampling.model(model, data, k, console = FALSE)

Arguments

model

model in R
data

data for the study of the model
k

number of re-samplings
console

logical, print output

Value

Model solution with resampling.

References

Efron, B., Tibshirani, R. (1993) An Introduction to the Boostrap. Chapman and Hall/CRC Phillip I. Good, (2001) Resampling Methods. Birkhauser. Boston . Basel . Berlin

See also

simulation.model

Examples


#example 1 Simple linear regression
library(agricolae)
data(clay)
model<-"ralstonia ~ days"
analysis<-resampling.model(model,clay,k=2,console=TRUE)
#> 
#> Resampling of the experiments
#> - - - - - - - - - - - - - - 
#> Proposed model: ralstonia ~ days 
#> ---
#> Resampling of the analysis of variancia for the proposed model
#> Determination of the P-Value by Resampling
#> Samples: 2 
#> 
#>           Df    Sum Sq   Mean Sq  F value       Pr(>F) Resampling
#> days       1  42.12812 42.128116 11.96848 0.0009456218          0
#> Residuals 67 235.83468  3.519921                                 
#> ---
#> 

#example 2 Analysis of variance: RCD
data(sweetpotato)
model<-"yield~virus"
analysis<-resampling.model(model,sweetpotato,k=2,console=TRUE)
#> 
#> Resampling of the experiments
#> - - - - - - - - - - - - - - 
#> Proposed model: yield~virus 
#> ---
#> Resampling of the analysis of variancia for the proposed model
#> Determination of the P-Value by Resampling
#> Samples: 2 
#> 
#>           Df    Sum Sq   Mean Sq  F value       Pr(>F) Resampling
#> virus      3 1170.2092 390.06972 17.34478 0.0007334172          0
#> Residuals  8  179.9133  22.48917                                 
#> ---
#> 

#example 3 Simple linear regression
data(Glycoalkaloids)
model<-"HPLC ~ spectrophotometer"
analysis<-resampling.model(model,Glycoalkaloids,k=2,console=TRUE)
#> 
#> Resampling of the experiments
#> - - - - - - - - - - - - - - 
#> Proposed model: HPLC ~ spectrophotometer 
#> ---
#> Resampling of the analysis of variancia for the proposed model
#> Determination of the P-Value by Resampling
#> Samples: 2 
#> 
#>                   Df     Sum Sq     Mean Sq  F value       Pr(>F) Resampling
#> spectrophotometer  1 449.506278 449.5062783 2527.078 5.042799e-25          0
#> Residuals         23   4.091146   0.1778759                                 
#> ---
#> 

#example 4 Factorial in RCD

data(potato)
potato[,1]<-as.factor(potato[,1])
potato[,2]<-as.factor(potato[,2])
model<-"cutting~variety + date + variety:date"
analysis<-resampling.model(model,potato,k=2,console=TRUE)
#> 
#> Resampling of the experiments
#> - - - - - - - - - - - - - - 
#> Proposed model: cutting~variety + date + variety:date 
#> ---
#> Resampling of the analysis of variancia for the proposed model
#> Determination of the P-Value by Resampling
#> Samples: 2 
#> 
#>              Df    Sum Sq   Mean Sq   F value     Pr(>F) Resampling
#> variety       1 25.086806 25.086806 7.2580377 0.01952218        0.0
#> date          2 13.891758  6.945879 2.0095604 0.17670768        0.5
#> variety:date  2  4.853025  2.426513 0.7020312 0.51483592        0.0
#> Residuals    12 41.477005  3.456417                                
#> ---
#>