MDopt: Best Model Discrimination (MD) Follow-Up Experiments

In ana-vela7/BsMD2: Bayes Screening and Model Discrimination

Description

Best follow-up experiments based on the MD criterion are suggested to discriminate between competing models.

Usage

MDopt(
  X,
  y,
  Xcand,
  nMod,
  p_mod,
  fac_mod,
  nFDes = 4,
  max_int = 3,
  g = 2,
  Iter = 20,
  nStart = 10,
  top = 10
)

Arguments

X	Matrix. Design matrix of the initial experiment.
y	Vector. Response vector of the initial experiment.
Xcand	Matrix. Candidate runs to be chosen for the follow-up design.
nMod	Integer. Number of competing models.
p_mod	Vector. Posterior probabilities of the competing models.
fac_mod	Matrix. Active factors in the competing models.
nFDes	Integer. Number of runs to consider in the follow-up experiment.
max_int	Integer. Maximum order of interactions in the models.
g	Numeric. Variance inflation factor for active effects.
Iter	Integer. Maximum number of iterations for each search.
nStart	Integer. Number of random starting designs.
top	Integer. Highest MD follow-up designs recorded.

Value

A list with all the input and output parameters.

X	Matrix. The design matrix.
y	Vector. The response vector.
Xcand	Matrix. Candidate runs to be chosen for the follow-up design.
Runs	Integer. Number of runs.
Fac	Integer. Number of factors.
nMod	Integer. Number of competing models.
p_mod	Vector. Posterior probabilities of the competing models.
fac_mod	Matrix. Active factors in the competing models.
nFDes	Integer. Number of runs to consider in the follow-up experiment.
max_int	Integer. Maximum order of the interactions in the models.
g	Numeric. Variance inflation factor for active effects.
Iter	Integer. Maximum number of iterations for each search.
nStart	Integer. Number of random starting designs.
top	Integer. Highest MD follow-up designs recorded.
MD	Data frame. Designs points and MD for top designs.
MDtop	Vector. MD for top designs.
DEStop	Data frame. Top design points.

References

Meyer, R. D., Steinberg, D. M. and Box, G. E. P. (1996). "Follow-Up Designs to Resolve Confounding in Multifactor Experiments (with discussion)". Technometrics, Vol. 38, No. 4, pp. 303-332.

Box, G. E. P and R. D. Meyer (1993). "Finding the Active Factors in Fractionated Screening Experiments". Journal of Quality Technology. Vol. 25. No. 2. pp. 94–105.

Examples

#Example 1
library(BsMD2)
data(BM93e3)
X <- as.matrix(BM93e3[1:16,c(1,2,4,6,9)]) #matriz de diseño inicial
y <- as.vector(BM93e3[1:16,10]) #vector de respuesta
p_mod <- c(0.2356,0.2356,0.2356,0.2356,0.0566) #probabilidad posterior de los 5 modelos
fac_mod <- matrix(c(2,1,1,1,1,3,3,2,2,2,4,4,3,4,3,0,0,0,0,4),nrow=5,
                  dimnames=list(1:5,c("f1","f2","f3","f4")))
Xcand <- matrix(c(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
                  -1,-1,-1,-1,1,1,1,1,-1,-1,-1,-1,1,1,1,1,
                  -1,-1,1,1,-1,-1,1,1,-1,-1,1,1,-1,-1,1,1,
                  -1,1,-1,1,-1,1,-1,1,-1,1,-1,1,-1,1,-1,1,
                  -1,1,1,-1,1,-1,-1,1,1,-1,-1,1,-1,1,1,-1),
                nrow=16,dimnames=list(1:16,c("blk","f1","f2","f3","f4"))
)
injectionMolding <- MDopt(X = X, y = y, Xcand = Xcand, nMod = 5, p_mod = p_mod, fac_mod = fac_mod,
nStart = 25)

#Example 2
data(M96e2,package="BsMD2")
X <- as.matrix(cbind(blk = rep(-1,8), M96e2[c(25,2,19,12,13,22,7,32), 1:5]))
y <- M96e2[c(25,2,19,12,13,22,7,32), 6]
pp <- BsProb1(X = X[,2:6], y = y, p = .25, gamma = .4, max_int = 3, max_fac = 5, top = 32)
p <- pp@p_mod
facs <- pp@fac_mod
Xcand <- as.matrix(cbind(blk = rep(+1,32), M96e2[,1:5]))
#e2 <- MDopt(X = X, y = y, Xcand = Xcand, nMod = 32, p_mod = p, fac_mod = facs, g = .4,
#Iter = 10, nStart = 25, top = 5)

ana-vela7/BsMD2 documentation built on Dec. 19, 2021, 2:32 a.m.