Title: | Tests of Homogeneity of Variances |
Version: | 0.1 |
Description: | Most common exact, asymptotic and resample based tests are provided for testing the homogeneity of variances of k normal distributions under normality. These tests are Barlett, Bhandary & Dai, Brown & Forsythe, Chang et al., Gokpinar & Gokpinar, Levene, Liu and Xu, Gokpinar. Also, a data generation function from multiple normal distribution is provided using any multiple normal parameters. Bartlett, M. S. (1937) <doi:10.1098/rspa.1937.0109> Bhandary, M., & Dai, H. (2008) <doi:10.1080/03610910802431011> Brown, M. B., & Forsythe, A. B. (1974).<doi:10.1080/01621459.1974.10482955> Chang, C. H., Pal, N., & Lin, J. J. (2017) <doi:10.1080/03610918.2016.1202277> Gokpinar E. & Gokpinar F. (2017) <doi:10.1080/03610918.2014.955110> Liu, X., & Xu, X. (2010) <doi:10.1016/j.spl.2010.05.017> Levene, H. (1960) https://cir.nii.ac.jp/crid/1573950400526848896 Gökpınar, E. (2020) <doi:10.1080/03610918.2020.1800037>. |
License: | GPL-2 |
Depends: | R (≥ 3.5.0), huxtable (≥ 5.4.0) |
Imports: | stats, graphics |
Encoding: | UTF-8 |
LazyData: | TRUE |
RoxygenNote: | 7.2.3 |
NeedsCompilation: | no |
Packaged: | 2023-02-11 16:54:31 UTC; macbook |
Author: | Fikri Gökpınar |
Maintainer: | Fikri Gökpınar <fikri@gazi.edu.tr> |
Repository: | CRAN |
Date/Publication: | 2023-02-13 09:30:02 UTC |
Brown-Forsythe Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups.
Usage
Brown_Forsythe(x1, x2, alfa = 0.05, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Brown, M. B., & Forsythe, A. B. (1974). Robust tests for the equality of variances. Journal of the American Statistical Association, 69(346), 364-367.
See Also
bdai
, Cat_GG
, Cat_LR
, genp
, slrt
, levene
,
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
Brown_Forsythe(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
Brown_Forsythe(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
Brown_Forsythe(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
Brown_Forsythe(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
Brown_Forsythe(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Computational Approach Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups using standartized likelihood ratio test.
Usage
Cat_GG(x1, x2, alfa = 0.05, m = 2000, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
m |
number of resampling. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Gokpinar, E., & Gokpinar, F. (2017). Testing equality of variances for several normal populations. Communications in Statistics-Simulation and Computation, 46(1), 38-52.
See Also
Brown_Forsythe
, bdai
, Cat_LR
, genp
, slrt
, levene
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
Cat_GG(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_GG(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_GG(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_GG(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_GG(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Computational Approach Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups using standartized likelihood ratio test.
Usage
Cat_LR(x1, x2, alfa = 0.05, m = 2000, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
m |
number of resampling. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Chang, C. H., Pal, N., & Lin, J. J. (2017). A revisit to test the equality of variances of several populations. Communications in Statistics-Simulation and Computation, 46(8), 6360-6384.
See Also
Brown_Forsythe
, Cat_GG
, bdai
, genp
, slrt
, levene
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
Cat_LR(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_LR(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_LR(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_LR(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
Cat_LR(x1,x2,alfa=0.10,m=5000,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Fleming and Harrington Data
Description
The data related to survival times of patients was collected from 4 hospitals, which was a part of the data by given Fleming and Harrington(1991). The data contain failure time of the patients.
Usage
data(FH_data)
Format
A dataframe with 21 rows 2 variables
- HospitalNo
Hospital No
- SurvivalTime
Survival Time of Patients
Source
T.R. Fleming and D.P. Harrington, Counting processes and survival analysis. Wiley Online Library, Vol. 8., 1991.
Examples
data("FH_data")
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
Bartlett Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups.
Usage
bart(x1, x2, alfa = 0.05, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Bartlett, M. S. (1937). "Properties of sufficiency and statistical tests". Proceedings of the Royal Statistical Society, Series A 160, 268–282 JSTOR.
See Also
levene
Brown_Forsythe
, Cat_GG
, Cat_LR
, genp
, slrt
, bdai
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
bart(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
bart(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
bart(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
bart(x1,x2,alfa=0.10,table=FALSE,graph="centerized")
readline(prompt = "Pause. Press <Enter> to continue...")
bart(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Bahandary-Dai Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups using Bahandary-Dai test.
Usage
bdai(x1, x2, alfa = 0.05, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Bhandary, M., & Dai, H. (2008). An alternative test for the equality of variances for several populations when the underlying distributions are normal. Communications in Statistics-Simulation and Computation, 38(1), 109-117.
See Also
Brown_Forsythe
, Cat_GG
, Cat_LR
, genp
, slrt
, levene
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
bdai(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
bdai(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
bdai(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
bdai(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
bdai(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Multiple Normal Distribution Data Generation
Description
This function generates data from multiple normal distribution.
Usage
datagen(n, mu, sigma, tn = 1)
Arguments
n |
Sample sizes of each group. n=c(n1,n2,...nk); for example: n=c(3, 4, 5). |
mu |
Mean of each group.mu=c(mu1,mu2,...muk); for example: mu=c(3, 4, 5). |
sigma |
Standard deviation of each group.sigma=c(sigma1,sigma2,...sigmak); for example: sigma=c(1, 2, 3). |
tn |
Trial number for all groups. Default of the parameter is 1. This parameter for use more than 1, is especially useful for resampling such as Monte Carlo, Parametric Bootstrap. |
Value
a data matrix with size (n1,n2,...nk) with group number 1,2,...k at first row and random number with mnean mu=(mu1,mu2,...muk) and standard deviation sigma=(sigma1,sigma2,...sigmak)
Examples
n=c(3, 4, 5)
mu=c(3, 4, 5)
sigma=c(3, 4, 5)
F=datagen(n,mu,sigma);muh=F[1];S2h=F[2];x=F[3]
muh
S2h
x
# Following example especially useful for simulation based tecnhiques
# such as Monte Carlo, Parametric Bootstrap and comparison studies
# by using simulation.
Fm=datagen(c(3, 4, 5),c(3, 4, 5),c(3, 4, 5),10);muhm=Fm[1];S2hm=Fm[2];xm=Fm[3]
muhm
S2hm
xm
Generalized p value Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups using generalized p value test.
Usage
genp(x1, x2, alfa = 0.05, m = 2000, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
m |
number of resampling. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Liu, X., & Xu, X. (2010). A new generalized p-value approach for testing the homogeneity of variances. Statistics & probability letters, 80(19-20), 1486-1491.
See Also
Brown_Forsythe
, Cat_GG
, Cat_LR
, bdai
, slrt
, levene
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
genp(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
genp(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
genp(x1,x2,alfa=0.10,m=5000)
readline(prompt = "Pause. Press <Enter> to continue...")
genp(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
genp(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
genp(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Levene Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups.
Usage
levene(x1, x2, alfa = 0.05, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Levene, H. (1960). Robust tests for equality of variances, p 278–292. Contributions to probability and statistics: essays in honor of Harold Hotelling. Stanford University Press, Palo Alto, CA.
See Also
Brown_Forsythe
, Cat_GG
, Cat_LR
, genp
, slrt
, bdai
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
levene(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
levene(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
levene(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
levene(x1,x2,alfa=0.10,table=FALSE,graph="raw")
readline(prompt = "Pause. Press <Enter> to continue...")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value
Standartized Likelihood Ratio Test for Homogeniety
Description
Tests the homogeniety of variances for more than two normal groups using standartized likelihood ratio test.
Usage
slrt(x1, x2, alfa = 0.05, table = TRUE, graph = "none")
Arguments
x1 |
a numeric matrix containing the values of groups. |
x2 |
numeric matrix containing the values of group numbers. |
alfa |
significance level of the test. Default number is 0.05. |
table |
a logical variable that indicates table will appear or not. Default is TRUE. |
graph |
box plot of groups of raw or centered data. |
Value
if table is TRUE, then it gives a detailed table, else it gives a vector of r value(r=1 when null hypothesis was rejected and r=0 when null hypothesis was accepted) p-value and test statistic value.
References
Gökpınar, E. (2020). Standardized likelihood ratio test for homogeneity of variance of several normal populations. Communications in Statistics-Simulation and Computation, 1-11.
See Also
Brown_Forsythe
, datagen
,levene
,
Cat_LR
, genp
Examples
data(FH_data)
x1=FH_data$SurvivalTime
x2=FH_data$HospitalNo
slrt(x1,x2)
readline(prompt = "Pause. Press <Enter> to continue...")
slrt(x1,x2,alfa=0.10)
readline(prompt = "Pause. Press <Enter> to continue...")
slrt(x1,x2,alfa=0.10,table=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
slrt(x1,x2,alfa=0.10,table=FALSE,graph=FALSE)
readline(prompt = "Pause. Press <Enter> to continue...")
slrt(x1,x2,alfa=0.10,table=FALSE,graph="none")
# ---THIS VERSION IS ESPECIALLY USEFUL FOR COMPARISON STUDIES BY SIMULATION---
# #first value of the vector is r value(r=1 when rejected and r=0 when accepted null hypothesis)
# second value of the vector is the p-value and third value is the tests statistic value