An open API service indexing awesome lists of open source software.

https://github.com/mightymetrika/npboottprm

Nonparametric Bootstrap Test with Pooled Resampling
https://github.com/mightymetrika/npboottprm

datascience nonparametric r statistics

Last synced: about 1 year ago
JSON representation

Nonparametric Bootstrap Test with Pooled Resampling

Awesome Lists containing this project

README

          

---
output: github_document
---

```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```

# npboottprm

The goal of npboottprm is to provide a robust tool for conducting nonparametric bootstrap tests with pooled resampling. These tests are ideal for small sample sizes and include the independent t-test, paired t-test, and F-test. The package employs methods presented in Dwivedi, Mallawaarachchi, and Alvarado (2017).

## Installation

You can install the released version of npboottprm from [CRAN](https://CRAN.R-project.org):

```{r eval=FALSE}
install.packages("npboottprm")
```

To install the development version of npboottprm from GitHub, use the [devtools](https://devtools.r-lib.org/) package:

```{r, eval=FALSE}
# install.packages("devtools")
devtools::install_github("mightymetrika/npboottprm")
```

## Nonparametric bootstrap t-test

The following example demonstrates how to use the nonparboot() function to conduct an independent t-test.

```{r}
library(npboottprm)

# Use the simulated data included in the package
print(data_t)

# Run the test
res_t <- nonparboot(data = data_t,
x = "x",
grp = "grp",
nboot = 1000,
test = "t",
conf.level = 0.95,
seed = 183)

# Print the results, excluding the bootstrap distributions
print(res_t[!names(res_t) %in%
c("bootstrap.stat.dist", "bootstrap.effect.dist")])
```

## Nonparametric bootstrap paired t-test

The following example demonstrates how to use the nonparboot() function to conduct a paired t-test.

```{r}
# Use the simulated data included in the package
print(data_pt)

# Run the test
res_pt <- nonparboot(data = data_pt,
x = "x",
y = "y",
nboot = 1000,
test = "pt",
conf.level = 0.95,
seed = 166)

# Print the results, excluding the bootstrap distributions
print(res_pt[!names(res_pt) %in%
c("bootstrap.stat.dist", "bootstrap.effect.dist")])
```

## Nonparametric bootstrap F-test

The following example demonstrates how to use the nonparboot() function to conduct an F-test.

```{r}
# Use the simulated data included in the package
print(data_f)

# Run the test
res_f <- nonparboot(data = data_f,
x = "x",
grp = "grp",
nboot = 1000,
test = "F",
conf.level = 0.95,
seed = 397)

# Print the results, excluding the bootstrap distributions
print(res_f[!names(res_f) %in%
c("bootstrap.stat.dist", "bootstrap.effect.dist")])
```

Please note that the examples provided here use simulated data included in the package. When using this package with your own data, replace data_t, data_pt, and data_f with your own data frames, and adjust the x, y, and grp parameters as needed.

## References

Dwivedi AK, Mallawaarachchi I, Alvarado LA (2017). "Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method." Statistics in Medicine, 36 (14), 2187-2205.