Help for package EMSaov

Type:

Package

Title:

The Analysis of Variance with EMS

Version:

2.3

Date:

2018-05-09

Author:

Eun-Kyung Lee, Hye-Min Choe

Maintainer:

Eun-Kyung Lee <lee.eunk@gmail.com>

Description:

Provides the analysis of variance table including the expected mean squares (EMS) for various types of experimental design. When some variables are random effects or we use special experimental design such as nested design, repeated-measures design, or split-plot design, it is not easy to find the appropriate test, especially denominator for F-statistic which depends on EMS.

License:

GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]

Imports:

shiny, graphics

RoxygenNote:

6.0.1

NeedsCompilation:

Packaged:

2018-05-09 00:51:40 UTC; EKLee

Repository:

CRAN

Date/Publication:

2018-05-09 03:54:09 UTC

Calculate ANOVA with approximate F value

Description

Calculate ANOVA with approximate F value

Usage

ApproxF(SS.table,approx.name)

Arguments

SS.table

result from EMSanova

approx.name

rowname in SS.table to calculate approximate F value for the test.

Examples

data(film)
anova.result<-EMSanova(thickness~Gate*Operator*Day,data=film,
                        type=c("F","R","R"))
anova.result                         
ApproxF(SS.table=anova.result,approx.name="Gate") 
EMSanova(thickness~Gate+Operator+Day,data=film,
          type=c("F","R","R"),
          approximate=TRUE)

Calculate ANOVA table with EMS

Description

Calculate ANOVA table with EMS for various experimental design - factorial design, nested design, mixed effect model, etc.

Usage

EMSanova(formula,data,type=NULL,nested=NULL,
                 level=NULL,approximate=FALSE)

Arguments

formula

model formula

data

data frame for ANOVA

type

the list of fixed/random for each factor. "F" for the fixed effect, "R" for the random effect

nested

the list of nested effect

level

list of model level

approximate

calculate approximated F for "TRUE"

Examples

data(baseball)
anova.result<-EMSanova(velocity~Group+Subject+test,data=baseball,
                 type=c("F","R","F"),
                 nested=c(NA,"Group",NA),
                 level=c(1,1,2))
anova.result

Shiny App for the analysis of variance in various experimental designs

Description

Shiny App for the analysis of variance in various experimental designs

Usage

EMSaovApp(nested.N=2)

Arguments

nested.N

number of factors of possible crossed design which can nest a factor

Examples

#EMSaovApp()

Pooling nonsignificant interactions to Residuals

Description

Pooling nonsignificant interactions to Residuals

Usage

PooledANOVA(SS.table,del.ID)

Arguments

SS.table

result from EMSanova

del.ID

id's to combine sum of squares. Use rownames of SS.table

Examples

data(film)
anova.result<-EMSanova(thickness~Gate*Operator*Day,data=film,
                        type=c("F","R","R"))
anova.result 
del.ID<-c("Gate:Day","Residuals")
PooledANOVA(anova.result,del.ID)

Nested factorial design of Measurement of velocity

Description

There was on a measurement of velocity of a baseball throw in meters per second. Three groups of subjects were involved, two being subjected to special experimental training and the third acting as a control with no special training. Each group has 7 subjects and each subject was given a pretest and posttest.

Usage

data("baseball")

Format

A data frame with 42 observations on the following 4 variables.

velocity: a numeric vector
test: a factor with levels Pre Post
Group: a factor with levels I II III
Subject: a numeric vector

References

Example 11.4 in Fundamental Concepts in the Design of Experiments (3rd ed.) - Charles R. Hicks

Examples

data(baseball)
## maybe str(baseball) ; plot(baseball) ...

Dry-film thickness

Description

Two days in a given month were randomly selected in which to run an experiment. three operators were selected at random from a large pool of available operators. The experiment consisted of measuring the dry-film thickness of varnish in mils for three different gate settings: 2, 4, and 6.

Usage

data("film")

Format

A data frame with 36 observations on the following 4 variables.

thickness: a numeric vector
Gate: a numeric vector
Operator: a factor with levels A B C
Day: a numeric vector

References

Fundamental Concepts in the Design of Experiments (3rd ed.) - Charles R. Hicks

Examples

data(film)
## maybe str(film) ; plot(film) ...

Split-split plot design of Curerate index

Description

A study of the cure rate index on some samples of rubber. Three laboratories, three temperatures and three types of mix were involved. Once a temperature was set, all three mixes were subjected to that temperature and then another temperature was set and again all three mixes were involved, finally the third temperature was set.

Usage

data("rubber")

Format

A data frame with 108 observations on the following 5 variables.

cure: a numeric vector
Rep: a factor with levels I II III IV
Lap: a numeric vector
Temp: a numeric vector
Mix: a factor with levels A B C

References

Fundamental Concepts in the Design of Experiments (3rd ed.) - Charles R. Hicks

Examples

data(rubber)
## maybe str(rubber) ; plot(rubber) ...