Package 'OneTwoSamples'

Title: Deal with One and Two (Normal) Samples
Description: We introduce an R function one_two_sample() which can deal with one and two (normal) samples, Ying-Ying Zhang, Yi Wei (2012) <doi:10.2991/asshm-13.2013.29>. For one normal sample x, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x. For two normal samples x and y, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x and y, respectively. It also reports interval estimation and test of hypothesis of mu1-mu2 (the difference of the means of x and y) and sigma1^2 / sigma2^2 (the ratio of the variances of x and y), tests whether x and y are from the same population, finds the correlation coefficient of x and y if x and y have the same length.
Authors: Frederic Bertrand [cre] , Ying-Ying Zhang (Robert) [aut]
Maintainer: Frederic Bertrand <[email protected]>
License: GPL (>= 2)
Version: 1.1-0
Built: 2025-01-29 04:07:40 UTC
Source: https://github.com/cran/OneTwoSamples

Help Index


Deal with One and Two (Normal) Samples

Description

In this package, we introduce an R function one_two_sample() which can deal with one and two (normal) samples, Ying-Ying Zhang, Yi Wei (2012), doi:10.2991/asshm-13.2013.29. For one normal sample x, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x. For two normal samples x and y, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x and y, respectively. It also reports interval estimation and test of hypothesis of mu1-mu2 (the difference of the means of x and y) and sigma1^2/sigma2^2 (the ratio of the variances of x and y), tests whether x and y are from the same population, finds the correlation coefficient of x and y if x and y have the same length.

Details

Package: OneTwoSamples
Type: Package
Version: 1.1-0
Date: 2023-03-22
License: GPL (>= 2)

The most important functions are: one_two_sample() and one_sample().

Author(s)

Ying-Ying Zhang (Robert)

Maintainer: Frederic Bertrand <[email protected]>

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

library("OneTwoSamples")

Compute various descriptive statistics

Description

Compute various descriptive statistics of x, such as mean, median, skewness, and kurtosis, etc.

Usage

data_outline(x)

Arguments

x

A numeric vector.

Value

A data.frame with variables:

N

The length.

Mean

The mean.

Var

The variance.

std_dev

Standard deviation.

Median

The median.

std_mean

The standard error of the sample mean.

CV

The coefficient of variation.

CSS

The corrected sum of squares.

USS

The uncorrected sum of squares.

R

The extreme difference.

R1

The half extreme difference, or the difference of upper quartile and lower quartile.

Skewness

The coefficient of skewness.

Kurtosis

The coefficient of kurtosis.

row.names

1.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
data_outline(x)

Show details of an object

Description

Show details of an object.

Usage

detail(x)

Arguments

x

Any R object to be tested.

Value

A list with components:

x

The argument x.

isS4

Logical, indicates whether x is an S4 object.

isObject

Logical, indicates whether x is an object, i.e., with a class attribute.

class

The class of x.

attributes

The attributes of x. Usually result$attributes is also a list.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

See Also

isS4, is.object, class, attributes

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
t = t.test(x); t
detail(t)

Two sided interval estimation of mu of one normal sample

Description

Compute the two sided interval estimation of mu of one normal sample when the population variance is known or unknown.

Usage

interval_estimate1(x, sigma = -1, alpha = 0.05)

Arguments

x

A numeric vector.

sigma

The standard deviation of the population. sigma>=0 indicates it is known, sigma<0 indicates it is unknown. Default to unknown standard deviation.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

mean

The sample mean.

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
interval_estimate1(x, sigma = 0.2)
interval_estimate1(x)

Two sided interval estimation of mu1-mu2 of two normal samples

Description

Compute the two sided interval estimation of mu1-mu2 of two normal samples when the population variances are known, unknown equal, or unknown unequal.

Usage

interval_estimate2(x, y, sigma = c(-1, -1), var.equal = FALSE, alpha = 0.05)

Arguments

x

A numeric vector.

y

A numeric vector.

sigma

A numeric vector of length 2, which contains the standard deviations of two populations. When the standard deviations are known, input it, then the function computes the interval endpoints using normal population; when the standard deviations are unknown, ignore it, now we need to consider whether the two populations have equal variances. See var.equal below.

var.equal

A logical variable indicating whether to treat the two variances as being equal. If TRUE then the pooled variance is used to estimate the variance otherwise the Welch (or Satterthwaite) approximation to the degrees of freedom is used.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

mean

The difference of sample means xb-yb.

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y

interval_estimate2(x, y, sigma = c(0.2, 0.3))
interval_estimate2(x, y, var.equal = TRUE)
interval_estimate2(x, y)

Two sided interval estimation of mu of one non-normal sample with large sample size

Description

Compute the two sided interval estimation of mu of one non-normal sample with large sample size when the population variance is known or unknown.

Usage

interval_estimate3(x, sigma = -1, alpha = 0.05)

Arguments

x

A numeric vector.

sigma

The standard deviation of the population. sigma>=0 indicates it is known, sigma<0 indicates it is unknown. Default to unknown standard deviation.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

mean

The sample mean.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x = rexp(50, 1/2); x
interval_estimate3(x)

Two sided or one sided interval estimation of mu of one normal sample

Description

Compute the two sided or one sided interval estimation of mu of one normal sample when the population variance is known or unknown.

Usage

interval_estimate4(x, sigma = -1, side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

sigma

The standard deviation of the population. sigma>=0 indicates it is known, sigma<0 indicates it is unknown. Default to unknown standard deviation.

side

A parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

mean

The sample mean.

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
interval_estimate4(x, sigma = 0.2, side = -1)
interval_estimate4(x, side = 1)

Two sided or one sided interval estimation of mu1-mu2 of two normal samples

Description

Compute the two sided or one sided interval estimation of mu1-mu2 of two normal samples when the population variances are known, unknown equal, or unknown unequal.

Usage

interval_estimate5(x, y, sigma = c(-1, -1), var.equal = FALSE, side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

y

A numeric vector.

sigma

A numeric vector of length 2, which contains the standard deviations of two populations. When the standard deviations are known, input it, then the function computes the interval endpoints using normal population; when the standard deviations are unknown, ignore it, now we need to consider whether the two populations have equal variances. See var.equal below.

var.equal

A logical variable indicating whether to treat the two variances as being equal. If TRUE then the pooled variance is used to estimate the variance otherwise the Welch (or Satterthwaite) approximation to the degrees of freedom is used.

side

A parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

mean

The difference of sample means xb-yb.

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y

interval_estimate5(x, y, sigma = c(0.2, 0.3), side = -1)
interval_estimate5(x, y, var.equal = TRUE)
interval_estimate5(x, y)

Two sided interval estimation of sigma^2 of one normal sample

Description

Compute the two sided interval estimation of sigma^2 of one normal sample when the population mean is known or unknown.

Usage

interval_var1(x, mu = Inf, alpha = 0.05)

Arguments

x

A numeric vector.

mu

The population mean. When it is known, input it, and the function computes the interval endpoints using a chi-square distribution with degree of freedom n. When it is unknown, ignore it, and the function computes the interval endpoints using a chi-square distribution with degree of freedom n-1.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

var

The estimate of the population variance. When the population mean mu is known, var = mean((x-mu)^2). When mu is unknown, var = var(x).

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
interval_var1(x, mu = 1)
interval_var1(x)

Two sided interval estimation of sigma1^2 / sigma2^2 of two normal samples

Description

Compute the two sided interval estimation of sigma1^2 / sigma2^2 of two normal samples when the population means are known or unknown.

Usage

interval_var2(x, y, mu = c(Inf, Inf), alpha = 0.05)

Arguments

x

A numeric vector.

y

A numeric vector.

mu

The population means. When it is known, input it, and the function computes the interval endpoints using an F distribution with degree of freedom (n1, n2). When it is unknown, ignore it, and the function computes the interval endpoints using an F distribution with degree of freedom (n1-1, n2-1).

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

rate

The estimate of the ratio of population variances, rate = Sx2/Sy2. When the population means mu is known, Sx2 = 1/n1*sum((x-mu[1])^2) and Sy2 = 1/n2*sum((y-mu[2])^2. When mu is unknown, Sx2 = var(x) and Sy2 = var(y).

df1

The first degree of freedom.

df2

The second degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y
interval_var2(x, y, mu = c(1,2))
interval_var2(x, y)

Two sided or one sided interval estimation of sigma^2 of one normal sample

Description

Compute the two sided or one sided interval estimation of sigma^2 of one normal sample when the population mean is known or unknown.

Usage

interval_var3(x, mu = Inf, side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

mu

The population mean. When it is known, input it, and the function computes the interval endpoints using a chi-square distribution with degree of freedom n. When it is unknown, ignore it, and the function computes the interval endpoints using a chi-square distribution with degree of freedom n-1.

side

A parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

var

The estimate of the population variance. When the population mean mu is known, var = mean((x-mu)^2). When mu is unknown, var = var(x).

df

The degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
interval_var3(x, mu = 1, side = -1)
interval_var3(x)

Two sided or one sided interval estimation of sigma1^2 / sigma2^2 of two normal samples

Description

Compute the two sided or one sided interval estimation of sigma1^2 / sigma2^2 of two normal samples when the population means are known or unknown.

Usage

interval_var4(x, y, mu = c(Inf, Inf), side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

y

A numeric vector.

mu

The population means. When it is known, input it, and the function computes the interval endpoints using an F distribution with degree of freedom (n1, n2). When it is unknown, ignore it, and the function computes the interval endpoints using an F distribution with degree of freedom (n1-1, n2-1).

side

A parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A data.frame with variables:

rate

The estimate of the ratio of population variances, rate = Sx2/Sy2. When the population means mu is known, Sx2 = 1/n1*sum((x-mu[1])^2) and Sy2 = 1/n2*sum((y-mu[2])^2. When mu is unknown, Sx2 = var(x) and Sy2 = var(y).

df1

The first degree of freedom.

df2

The second degree of freedom.

a

The confidence lower limit.

b

The confidence upper limit.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y
interval_var4(x, y, mu = c(1,2), side = -1)
interval_var4(x, y)

Two sided or one sided test of hypothesis of mu of one normal sample

Description

Compute the two sided or one sided test of hypothesis of mu of one normal sample when the population variance is known or unknown.

Usage

mean_test1(x, mu = 0, sigma = -1, side = 0)

Arguments

x

A numeric vector.

mu

mu is mu0 in the null hypothesis. Default is 0, i.e., H0: mu = 0.

sigma

The standard deviation of the population. sigma>=0 indicates it is known, sigma<0 indicates it is unknown. Default to unknown standard deviation.

side

A parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: mu != mu0; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: mu < mu0; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: mu > mu0.

Value

A data.frame with variables:

mean

The sample mean.

df

The degree of freedom.

statistic

The statistic, when sigma>=0, statistic = Z = (xb-mu)/(sigma/sqrt(n)); when sigma<0, statistic = T = (xb-mu)/(sd(x)/sqrt(n)).

p_value

The P value.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
mean_test1(x, mu = 1, sigma = 0.2, side = 1)
mean_test1(x, mu = 1)

Two sided or one sided test of hypothesis of mu1 and mu2 of two normal samples

Description

Compute the two sided or one sided test of hypothesis of mu1 and mu2 of two normal samples when the population variances are known, unknown equal, or unknown unequal.

Usage

mean_test2(x, y, sigma = c(-1, -1), var.equal = FALSE, side = 0)

Arguments

x

A numeric vector.

y

A numeric vector.

sigma

A numeric vector of length 2, which contains the standard deviations of two populations. When the standard deviations are known, input it, then the function computes the interval endpoints using normal population; when the standard deviations are unknown, ignore it, now we need to consider whether the two populations have equal variances. See var.equal below.

var.equal

A logical variable indicating whether to treat the two variances as being equal. If TRUE then the pooled variance is used to estimate the variance otherwise the Welch (or Satterthwaite) approximation to the degrees of freedom is used.

side

A parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: mu1 != mu2; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: mu1 < mu2; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: mu1 > mu2.

Value

A data.frame with variables:

mean

The difference of sample means xb-yb.

df

The degree of freedom.

statistic

The statistic, when all(sigma>=0), statistic = Z; otherwise, statistic = T.

p_value

The P value.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y
mean_test2(x, y, sigma = c(0.2, 0.3), side = 1)
mean_test2(x, y, var.equal = TRUE, side = 1)
mean_test2(x, y, side = 1)

Deal with one (normal) sample

Description

Deal with one sample x, especially normal. Report descriptive statistics, plot, interval estimation and test of hypothesis of x.

Usage

one_sample(x, mu = Inf, sigma = -1, side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

mu

mu plays two roles.

In two sided or one sided interval estimation (or test of hypothesis) of sigma^2 of one normal sample, mu is the population mean. When it is known, input it, and the function computes the interval endpoints (or chi-square statistic) using a chi-square distribution with degree of freedom n. When it is unknown, ignore it (the default), and the function computes the interval endpoints (or chi-square statistic) using a chi-square distribution with degree of freedom n-1.

In two sided or one sided test of hypothesis of mu of one normal sample, mu is mu0 in the null hypothesis, and mu0 = if (mu < Inf) mu else 0.

sigma

sigma plays two roles.

In two sided or one sided interval estimation (or test of hypothesis) of mu of one normal sample, sigma is the standard deviation of the population. sigma>=0 indicates it is known, and the function computes the interval endpoints (or Z statistic) using a standard normal distribution. sigma<0 indicates it is unknown, and the function computes the interval endpoints (or T statistic) using a t distribution with degree of freedom n-1. Default to unknown standard deviation.

In two sided or one sided test of hypothesis of sigma^2 of one normal sample, sigma is sigma0 in the null hypothesis. Default is 1, i.e., H0: sigma^2 = 1.

side

side plays two roles and is used in four places.

In two sided or one sided interval estimation of mu of one normal sample, side is a parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

In two sided or one sided interval estimation of sigma^2 of one normal sample, side is a parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1; when computing the one sided lower limit, input side = 1; when computing the two sided limits, input side = 0 (default).

In two sided or one sided test of hypothesis of mu of one normal sample, side is a parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: mu != mu0; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: mu < mu0; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: mu > mu0.

In two sided or one sided test of hypothesis of sigma^2 of one normal sample, side is a parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: sigma^2 != sigma0^2; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: sigma^2 < sigma0^2; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: sigma^2 > sigma0^2.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A list with the following components:

mu_interval

It contains the results of interval estimation of mu.

mu_hypothesis

It contains the results of test of hypothesis of mu.

sigma_interval

It contains the results of interval estimation of sigma.

sigma_hypothesis

It contains the results of test of hypothesis of sigma.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
one_sample(x, mu = 1, sigma = 0.2, side = 1)
one_sample(x, sigma = 0.2, side = 1)
one_sample(x, mu = 1, side = 1)
one_sample(x)

Deal with one and two (normal) samples

Description

Deal with one and two (normal) samples. For one normal sample x, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x. For two normal samples x and y, the function reports descriptive statistics, plot, interval estimation and test of hypothesis of x and y, respectively. It also reports interval estimation and test of hypothesis of mu1-mu2 (the difference of the means of x and y) and sigma1^2/sigma2^2 (the ratio of the variances of x and y), tests whether x and y are from the same population, finds the correlation coefficient of x and y if x and y have the same length.

Usage

one_two_sample(x, y = NULL, mu = c(Inf, Inf), sigma = c(-1, -1), 
               var.equal = FALSE, ratio = 1, side = 0, alpha = 0.05)

Arguments

x

A numeric vector.

y

A numeric vector.

mu

If y = NULL, i.e., there is only one sample. See the argument mu in one_sample. For two normal samples x and y, mu plays one role: the population means. However, mu is used in two places: one is the two sided or one sided interval estimation of sigma1^2 / sigma2^2 of two normal samples, another is the two sided or one sided test of hypothesis of sigma1^2 and sigma2^2 of two normal samples. When mu is known, input it, and the function computes the interval endpoints (or the F value) using an F distribution with degree of freedom (n1, n2). When it is unknown, ignore it, and the function computes the interval endpoints (or the F value) using an F distribution with degree of freedom (n1-1, n2-1).

sigma

If y = NULL, i.e., there is only one sample. See the argument sigma in one_sample. For two normal samples x and y, sigma plays one role: the population standard deviations. However, sigma is used in two places: one is the two sided or one sided interval estimation of mu1-mu2 of two normal samples, another is the two sided or one sided test of hypothesis of mu1 and mu2 of two normal samples. When the standard deviations are known, input it, then the function computes the interval endpoints using normal population; when the standard deviations are unknown, ignore it, now we need to consider whether the two populations have equal variances. See var.equal below.

var.equal

A logical variable indicating whether to treat the two variances as being equal. If TRUE then the pooled variance is used to estimate the variance otherwise the Welch (or Satterthwaite) approximation to the degrees of freedom is used.

ratio

The hypothesized ratio of the population variances of x and y. It is used in var.test(x, y, ratio = ratio, ...), i.e., when computing the interval estimation and test of hypothesis of sigma1^2 / sigma2^2 when mu1 or mu2 is unknown.

side

If y = NULL, i.e., there is only one sample. See the argument side in one_sample. For two normal samples x and y, sigma is used in four places: interval estimation of mu1-mu2, test of hypothesis of mu1 and mu2, interval estimation of sigma1^2 / sigma2^2, test of hypothesis of sigma1^2 and sigma2^2. In interval estimation of mu1-mu2 or sigma1^2 / sigma2^2, side is a parameter used to control whether to compute two sided or one sided interval estimation. When computing the one sided upper limit, input side = -1 (or a number < 0); when computing the one sided lower limit, input side = 1 (or a number > 0); when computing the two sided limits, input side = 0 (default). In test of hypothesis of mu1 and mu2 or sigma1^2 and sigma2^2, side is a parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: mu1 != mu2 or H1: sigma1^2 != sigma2^2; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: mu1 < mu2 or H1: sigma1^2 < sigma2^2; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: mu1 > mu2 or H1: sigma1^2 > sigma2^2.

alpha

The significance level, a real number in [0, 1]. Default to 0.05. 1-alpha is the degree of confidence.

Value

A list with the following components:

one_sample_x

It contains the results by one_sample(x, ...).

one_sample_y

It contains the results by one_sample(y, ...).

mu1_mu2_interval

It contains the results of interval estimation of mu1-mu2.

mu1_mu2_hypothesis

It contains the results of test of hypothesis of mu1-mu2.

sigma_ratio_interval

It contains the results of interval estimation of sigma1^2 / sigma2^2.

sigma_ratio_hypothesis

It contains the results of test of hypothesis of sigma1^2 / sigma2^2.

res.ks

It contains the results of ks.test(x,y).

res.binom

It contains the results of binom.test(sum(x<y), length(x)).

res.wilcox

It contains the results of wilcox.test(x, y, ...).

cor.pearson

It contains the results of cor.test(x, y, method = "pearson", ...).

cor.kendall

It contains the results of cor.test(x, y, method = "kendall", ...).

cor.spearman

It contains the results of cor.test(x, y, method = "spearman", ...).

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

## One sample
x=rnorm(10, mean = 1, sd = 0.2); x

## one_sample(x, ...) == one_two_sample(x, ...)
one_sample(x, mu = 1, sigma = 0.2, side = 1)
one_two_sample(x, mu = 1, sigma = 0.2, side = 1)

one_sample(x, sigma = 0.2, side = 1)
one_two_sample(x, sigma = 0.2, side = 1)

one_sample(x, mu = 1, side = 1)
one_two_sample(x, mu = 1, side = 1)

one_sample(x)
one_two_sample(x)

## Two samples
set.seed(1)
x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y
y2=rnorm(20, mean = 2, sd = 0.2); y2

## sigma1, sigma2 known; mu1, mu2 known
one_two_sample(x, y, sigma = c(0.2, 0.3), mu = c(1, 2))

## sigma1 = sigma2 unknown; mu1, mu2 known
one_two_sample(x, y2, var.equal = TRUE, mu = c(1, 2))

## sigma1 != sigma2 unknown; mu1, mu2 known
one_two_sample(x, y, mu = c(1, 2))

## sigma1, sigma2 known; mu1, mu2 unknown
one_two_sample(x, y, sigma = c(0.2, 0.3))

## sigma1 = sigma2 unknown; mu1, mu2 unknown
one_two_sample(x, y2, var.equal = TRUE)

## sigma1 != sigma2 unknown; mu1, mu2 unknown
one_two_sample(x, y)

Compute the P value

Description

Compute the P value of a cumulative distribution function (cdf).

Usage

p_value(cdf, x, paramet = numeric(0), side = 0)

Arguments

cdf

The cumulative distribution function. For normal distribution, cdf = pnorm.

x

A given value to compute the P value.

paramet

The parameter of the corresponding distribution. For normal distribution, paramet = c(mu, sigma).

side

A parameter indicating whether to compute one sided or two sided P value. When inputting side = -1 (or a number < 0), the function computes a left side P value; when inputting side = 1 (or a number > 0), the function computes a right side P value; when inputting side = 0 (default), the function computes a two sided P value.

Value

The P value.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

p_value(pnorm, x = 0, side = 1)
p_value(pt, x = 0, paramet = 5, side = 1)

Two sided or one sided test of hypothesis of sigma^2 of one normal sample

Description

Compute the two sided or one sided test of hypothesis of sigma^2 of one normal sample when the population mean is known or unknown.

Usage

var_test1(x, sigma2 = 1, mu = Inf, side = 0)

Arguments

x

A numeric vector.

sigma2

sigma2 is sigma0^2 in the null hypothesis. Default is 1, i.e., H0: sigma^2 = 1.

mu

The population mean. mu < Inf indicates it is known, mu == Inf indicates it is unknown. Default to unknown population mean.

side

A parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: sigma^2 != sigma0^2; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: sigma^2 < sigma0^2; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: sigma^2 > sigma0^2.

Value

A data.frame with variables:

var

The estimate of the population variance. When the population mean mu is known, var = mean((x-mu)^2). When mu is unknown, var = var(x).

df

The degree of freedom.

chisq2

The chisquare statistic.

p_value

The P value.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
var_test1(x, sigma2 = 0.2^2, mu = 1, side = 1)
var_test1(x, sigma2 = 0.2^2, side = 1)

Two sided or one sided test of hypothesis of sigma1^2 and sigma2^2 of two normal samples

Description

Compute the two sided or one sided test of hypothesis of sigma1^2 and sigma2^2 of two normal samples when the population means are known or unknown.

Usage

var_test2(x, y, mu = c(Inf, Inf), side = 0)

Arguments

x

A numeric vector.

y

A numeric vector.

mu

The population means. When it is known, input it, and the function computes the F value using an F distribution with degree of freedom (n1, n2). When it is unknown, ignore it, and the function computes the F value using an F distribution with degree of freedom (n1-1, n2-1).

side

A parameter used to control two sided or one sided test of hypothesis. When inputting side = 0 (default), the function computes two sided test of hypothesis, and H1: sigma1^2 != sigma2^2; when inputting side = -1 (or a number < 0), the function computes one sided test of hypothesis, and H1: sigma1^2 < sigma2^2; when inputting side = 1 (or a number > 0), the function computes one sided test of hypothesis, and H1: sigma1^2 > sigma2^2.

Value

A data.frame with variables:

rate

The estimate of the ratio of population variances, rate = Sx2/Sy2. When the population means mu is known, Sx2 = 1/n1*sum((x-mu[1])^2) and Sy2 = 1/n2*sum((y-mu[2])^2. When mu is unknown, Sx2 = var(x) and Sy2 = var(y).

df1

The first degree of freedom.

df2

The second degree of freedom.

F

The F statistic.

p_value

The P value.

Author(s)

Ying-Ying Zhang (Robert) [email protected]

References

Zhang, Y. Y., Wei, Y. (2013), One and two samples using only an R funtion, doi:10.2991/asshm-13.2013.29.

Examples

x=rnorm(10, mean = 1, sd = 0.2); x
y=rnorm(20, mean = 2, sd = 0.3); y
var_test2(x, y, mu = c(1, 2), side = 1)
var_test2(x, y, side = 1)