SK.nest {ScottKnott}R Documentation

The ScottKnott Clustering Algoritm for Factorial, Split-plot and Split-SPlit plot Experiments

Description

These are methods for objects of class vector, matrix or data.frame joined as default, aov and aovlist for factorial, split-plot and split-split-plot experiments.

Usage

  ## Default S3 method:
  SK.nest(x, y=NULL, model, which, error, fl2, fl3=0, sig.level=.05, ...)
  ## S3 method for class 'aov':
  SK.nest(x, which, fl2, fl3=0, sig.level=.05, ...)
  ## S3 method for class 'aovlist':
  SK.nest(x, which, error, fl2, fl3=0, sig.level=.05, ...)

Arguments

x A design matrix, data.frame or an aov object.
y A vector of response variable. It is necessary to inform this parameter only if x represent the design matrix without the response variable in it
which The name of the treatment to be used in the comparison. The name must be inside quoting marks.
model If x is a data.frame object, the model to be used in the aov must be specified.
fl2 A vector of length 1 giving the level of the second factor in nesting order tested.
fl3 A vector of length 1 giving the level of the third factor in nesting order tested.
error The error to be considered, only in case of split-plots experiments.
sig.level Level of Significance used in the SK algorithm to create the groups of means. The default value is 0.05.
... Potential further arguments (require by generic).

Details

The function SK.nest returns an object of class SK.nest containing the groups of means plus other necessary variables for summary and plot.

The generic functions summary and plot are used to obtain and print a summary and a plot of the results.

Value

The function SK.nest returns a list of the class SK.nest with the slots:

av A list storing the result of aov.
groups A vector of length equal the number of treatments marking the groups generated.
nms A vector of the labels of the treatments.
ord A vector which keeps the position of the means of the treatments in decreasing order.
means A vector which keeps the means of the treatments in decreasing order.
sig.level A vector of length 1 giving the level of significance of the test.
mnumber A vector of length 1 giving the number of treatments.
r A vector of length 1 giving the number of replicates.
which The name of the factor whose levels were tested.
fl2 A vector of length 1 giving the level of the second factor in nesting order tested.
fl3 A vector of length 1 giving the level of the third factor in nesting order tested.

Author(s)

Enio Jelihovschi (eniojelihovs@gmail.com)
Jose Claudio Faria (joseclaudio.faria@gmail.com)
Sergio Oliveira (solive@uesc.br)

References

Ramalho MAP, Ferreira DF, Oliveira AC 2000. Experimentacao em Genetica e Melhoramento de Plantas. Editora UFLA.

Scott RJ, Knott M 1974. A cluster analysis method for grouping mans in the analysis of variance. Biometrics, 30, 507-512.

Examples

  ##
  ## Example: Split-split-plot Experiment (SSPE)
  ##

  A   <- gl(3, 60, 180, 1:3)
  C   <- rep(gl(3, 20, 60, 1:3), 3)
  D   <- rep(gl(5, 4, 20, 1:5), 9)
  blo <- rep(gl(4, 1, 4, 1:4), 45)

  y <- c(
    3029, 3857, 2448, 2448, 2438, 3086, 3771, 4657, 3543, 4314, 3888, 3945,
    4318, 4514, 4497, 4215, 4610, 4590, 4499, 4545, 3448, 3600, 4267, 3895,
    3533, 5048, 3467, 4095, 3770, 3840, 3699, 3801, 3975, 3840, 3725, 3952,
    4010, 3880, 3785, 3903, 3706, 3815, 3795, 3900, 3870, 4003, 3987, 3938,
    4040, 3879, 3777, 4218, 3888, 3964, 3800, 3715, 3815, 3793, 3905, 3918,
    3074, 3852, 2483, 2453, 2553, 3121, 3786, 4672, 3558, 4429, 3913, 3860,
    4233, 4538, 4612, 4240, 4635, 4615, 4524, 4570, 3473, 3625, 4292, 3920,
    3558, 5193, 3492, 4210, 3780, 3850, 3709, 3723, 3895, 3940, 3735, 3962,
    4190, 3890, 3795, 3813, 3716, 3825, 3825, 3985, 3862, 3898, 3982, 3993,
    4105, 3874, 3772, 4213, 3896, 3982, 3808, 3833, 3893, 3871, 3923, 3936,
    3256, 3804, 2745, 2475, 2565, 3163, 3798, 4684, 3770, 4341, 3815, 3972,
    4355, 4541, 4411, 4229, 4624, 4504, 4713, 4509, 3462, 3614, 4281, 3909,
    3547, 4962, 3481, 4109, 3874, 3654, 3713, 3828, 4092, 3867, 3752, 3979,
    3837, 3707, 3872, 3730, 3833, 3942, 3722, 3912, 3882, 4015, 3989, 3890,
    4052, 3891, 3765, 4197, 3899, 3905, 3911, 3726, 3876, 3804, 3916, 3819)

  dm  <- data.frame(A, C, D, blo) # Design matrix (a data.frame object)
  dfm <- data.frame(A, C, D, blo, y)

  ## PARAMETERS ARE THE DESIGN MATRIX AND THE RESPONSE VARIABLE
  ## MAIN FACTOR ANALYSIS
  ## Main factor = A
  sk1 <- SK(dm, y, model='y ~ blo + D*C*A + Error(blo/A/C)', which='A',
            error='blo:A')
  summary(sk1)
  plot(sk1)

  # Main factor = C
  sk2 <- SK(dm, y, model='y ~ blo + D*C*A + Error(blo/A/C)', which='C',
            error='blo:A:C', sig.level=0.025)
  summary(sk2)
  plot(sk2, title='C', col=rainbow(3))

  # Main factor = D
  sk3 <- SK(dm, y, model='y ~ blo + D*C*A + Error(blo/A/C)', which='D',
            error='Within', sig.level=0.1)
  summary(sk3)
  plot(sk3, title='D', col=heat.colors(3))

  ## NESTED ANALYSIS
  ## Nested C/A=1 -> SK.nest
  sk4 <- SK.nest(dm, y, model='y ~ blo + D*C*A + Error(blo/A/C)', which='C:A',
                 error='blo:A:C', fl2=1)
  summary(sk4)
  plot(sk4, title='C/A=1', col='darkgray')

[Package ScottKnott version 1.0.0 Index]