Package 'epr'

Title: Easy Polynomial Regression
Description: Performs analysis of polynomial regression in simple designs with quantitative treatments.
Authors: Emmanuel Arnhold
Maintainer: Emmanuel Arnhold <[email protected]>
License: GPL-2
Version: 3.0
Built: 2024-12-17 06:53:48 UTC
Source: CRAN

Help Index


Easy Polynomial Regression

Description

Performs analysis of polynomial regression in simple designs with quantitative treatments.

Details

Package: epr
Type: Package
Version: 3.0
Date: 2017-11-14
License: GPL-2

Author(s)

Emmanuel Arnhold <[email protected]>

References

KAPS, M. and LAMBERSON, W. R. Biostatistics for Animal Science: an introductory text. 2nd Edition. CABI Publishing, Wallingford, Oxfordshire, UK, 2009. 504p.

SAMPAIO, I. B. M. Estatistica aplicada a experimentacao animal. 3nd Edition. Belo Horizonte: Editora FEPMVZ, Fundacao de Ensino e Pesquisa em Medicina Veterinaria e Zootecnia, 2010. 264p.

Examples

# analysis in completely randomized design
data(data1)
r1=pr2(data1)
names(r1)
r1
r1[1]

pr1(data1)

# analysis in randomized block design
data(data2)
r2=pr2(data2, design=2)
r2


# analysis in latin square design
data(data3)
r3=pr2(data3, design=3)
r3

# analysis in several latin squares
data(data4)
r4=pr2(data4, design=4)
r4

Analysis of bronken line regression

Description

The function performs analysis of broken line regression.

Usage

bl(data, xlab="Explanatory Variable", ylab="Response Variable", position=1)

Arguments

data

data is a data.frame

The first column should contain the treatments (explanatory variable) and the second column the response variable

xlab

name of explanatory variable

ylab

name of response variable

position

position of equation in the graph

top=1

bottomright=2

bottom=3

bottomleft=4

left=5

topleft=6 (default)

topright=7

right=8

center=9

Value

Returns coefficients of the models, t test for coefficients, R squared, adjusted R squared, AIC and BIC, normality test and residuals.

Author(s)

Emmanuel Arnhold <[email protected]>

See Also

lm, ea1(easyanova package), pr2, regplot

Examples

x=c(0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08,0.09,0.10)
y=c(5.5,4,3.2,2.1,1,0.1,1.6,2.2,3,5)
y=y/100
data=data.frame(x,y)

### bl(data)

data1: Sampaio (2010): page 134

Description

Quantitative treatments in completely randomized design.

Usage

data(data1)

Format

A data frame with 24 observations on the following 2 variables.

treatment

a numeric vector

gain

a numeric vector

References

SAMPAIO, I. B. M. Estatistica aplicada a experimentacao animal. 3nd Edition. Belo Horizonte: Editora FEPMVZ, Fundacao de Ensino e Pesquisa em Medicina Veterinaria e Zootecnia, 2010. 264p.

Examples

data(data1)
summary(data1)

data2: Kaps and Lamberson (2009): page 434

Description

Quantitative treatments in randomizad block design.

Usage

data(data2)

Format

A data frame with 25 observations on the following 3 variables.

protein_level

a numeric vector

litter

a factor with levels l1 l2 l3 l4 l5

feed_conversion

a numeric vector

References

KAPS, M. and LAMBERSON, W. R. Biostatistics for Animal Science: an introductory text. 2nd Edition. CABI Publishing, Wallingford, Oxfordshire, UK, 2009. 504p.

Examples

data(data2)
summary(data2)

data3: fictional example

Description

Quantitative treatments in latin square design.

Usage

data(data3)

Format

A data frame with 25 observations on the following 4 variables.

treatment

a numeric vector

animal

a factor with levels a1 a2 a3 a4 a5

period

a factor with levels p1 p2 p3 p4 p5

milk_fat

a numeric vector

Examples

data(data3)
summary(data3)

data4: fictional example

Description

Quantitative treatments in several latin squares design.

Usage

data(data4)

Format

A data frame with 50 observations on the following 5 variables.

treatment

a numeric vector

square

a numeric vector

animal

a factor with levels a1 a2 a3 a4 a5

period

a factor with levels p1 p2 p3 p4 p5

milk_fat

a numeric vector

Examples

data(data4)
summary(data4)

data5: fictional example

Description

Quantitative treatments and three response variable.

Usage

data(data5)

Format

A data frame with 24 observations on the following 4 variables.

treatments

a numeric vector

variable1

a numeric vector

variable2

a numeric vector

variable3

a numeric vector

Examples

data(data5)
summary(data5)

Analysis of polynomial regression

Description

The function performs analysis of polynomial regression in simple designs with quantitative treatments. The function also performs with randon factor in mixed models.

Usage

pr1(data, mixed = FALSE, digits = 6)

Arguments

data

data is a data.frame

The first column should contain the treatments (explanatory variable) and the remaining columns the response variables (fixed model).

The first column should contain the treatments (explanatory variable), second colunm should contais de random variable and the remaining columns the response variables (mixed model).

mixed

FALSE = fixed model

TRUE = mixed model

digits

6 = defalt (number of digits)

Value

Returns coefficients of the models, t test for coefficients, R squared, adjusted R squared, AIC, BIC and the maximum (or minimum) values of y and critical point of x, residuals and normality test.

Author(s)

Emmanuel Arnhold <[email protected]>

See Also

lm, ea1(easyanova package), pr2, regplot

Examples

# data
data(data5)

# linear and quadratic models
results1=pr1(data5)
results1

# analysis in completely randomized design
data(data1)
r1=pr2(data1)
names(r1)
r1
r1[1]

pr1(data1)

# analysis in randomized block design
data(data2)
r2=pr2(data2, design=2)
r2

pr1(data2, mixed=TRUE)

Analysis of polynomial regression

Description

The function performs analysis of polynomial regression in simple designs with quantitative treatments. This function performs analysis the lack of fit .

Usage

pr2(data, design = 1, list = FALSE, type = 2)

Arguments

data

data is a data.frame

data frame with two columns, treatments and response (completely randomized design)

data frame with three columns, treatments, blocks and response (randomized block design)

data frame with four columns, treatments, rows, cols and response (latin square design)

data frame with five columns, treatments, square, rows, cols and response (several latin squares)

design

1 = completely randomized design

2 = randomized block design

3 = latin square design

4 = several latin squares

list

FALSE = a single response variable

TRUE = multivariable response

type

type is form of obtain sum of squares

1 = a sequential sum of squares

2 = a partial sum of squares

Details

The response and the treatments must be numeric. Other variables can be numeric or factors.

Value

Returns analysis of variance, models, t test for coefficients and R squared and adjusted R squared.

Author(s)

Emmanuel Arnhold <[email protected]>

References

KAPS, M. and LAMBERSON, W. R. Biostatistics for Animal Science: an introductory text. 2nd Edition. CABI Publishing, Wallingford, Oxfordshire, UK, 2009. 504p.

SAMPAIO, I. B. M. Estatistica aplicada a experimentacao animal. 3nd Edition. Belo Horizonte: Editora FEPMVZ, Fundacao de Ensino e Pesquisa em Medicina Veterinaria e Zootecnia, 2010. 264p.

See Also

lm, lme(package nlme), ea1(package easyanova), pr1, regplot

Examples

# analysis in completely randomized design
data(data1)
r1=pr2(data1)
names(r1)
r1
r1[1]

# analysis in randomized block design
data(data2)
r2=pr2(data2, design=2)
r2

# analysis in latin square design
data(data3)
r3=pr2(data3, design=3)
r3

# analysis in several latin squares
data(data4)
r4=pr2(data4, design=4)
r4

# data
treatments=rep(c(0.5,1,1.5,2,2.5,3), c(3,3,3,3,3,3))
r1=rnorm(18,60,3)
r2=r1*1:18
r3=r1*18:1
r4=r1*c(c(1:10),10,10,10,10,10,10,10,10)
data6=data.frame(treatments,r1,r2,r3, r4)

# use the argument list = TRUE
pr2(data6, design=1, list=TRUE)

Tests for model identity and parameter

Description

The function performs tests of parameters and models.

Usage

r.test(data, digits=6)

Arguments

data

data is a data.frame The first column should contain the x (explanatory variable) second treatments and the remaining columns the response variables.

digits

number of digits (defalt = 6)

Value

Returns coefficients of the models, t test for coefficients and tests for parameters and models.

Author(s)

Emmanuel Arnhold <[email protected]>

See Also

lm, ea1(easyanova package), pr2, regplot

Examples

x=c(1,1,1,2,2,2,3,3,3,4,4,4)
y=c(5,5.3,6,8,8.9,12,14,18,25,25,29,32)
t=c("a1","a2","a3","a1","a2","a3","a1","a2","a3","a1","a2","a3")
data=data.frame(x,t,y)

r.test(data)

Graphics of the regression

Description

The function generates the scatter plot with the regression equation.

Usage

regplot(data, xlab="Explanatory Variable", ylab="Response Variable", 
position=6, mean=TRUE, digits=4)

Arguments

data

data is a data.frame

the first column contain the explanatory variable

the others columns contain the responses variables

xlab

name of variable x

ylab

name of variable y

position

position of equation in the graph

top=1

bottomright=2

bottom=3

bottomleft=4

left=5

topleft=6 (default)

topright=7

right=8

center=9

mean

TRUE = scatter plots with averages (default)

FALSE = scatter plots with all data

digits

number of digits

Value

The function generates the scatter plot with the regression equation.

Author(s)

Emmanuel Arnhold <[email protected]>

See Also

lm, lme, ea1(easyanova package), pr2, pr2, dplot(ds package)

Examples

# data
data(data5)

d1=data5[,c(1,2)]
regplot(d1, position=8)

d2=data5[,c(1,3)]
regplot(d2, position=8)

d3=data5[,c(1,4)]
regplot(d3, position=8)