Package 'npboottprmFBar'

Nonparametric Bootstrap Test with Pooled Resampling for Informative Hypothesis Testing

Description

Performs a nonparametric bootstrap test with pooled resampling based on the methods described by Dwivedi et al. (2017) and using the restriktor::iht function as proposed by Vanbrabant and Rosseel (2020).

Usage

bootFbar(
  data,
  formula,
  grp,
  constraints,
  nboot = 1000,
  conf.level = 0.95,
  seed = NULL,
  na_rm = FALSE
)

Arguments

data	A data frame containing the dataset for analysis.
formula	An object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted.
grp	A string specifying the grouping variable in the data.
constraints	A matrix or data frame of constraints for the hypothesis test.
nboot	An integer indicating the number of bootstrap resamples (default is 1000).
conf.level	A numeric value specifying the confidence level for the interval (default is 0.95).
seed	An optional integer setting the seed for random number generation (default is NULL).
na_rm	A logical value indicating whether NA values should be removed (default is FALSE).

Value

A list containing the following components:

-pvalueA, pvalueB, pvalueF: P-values for the different test types.

-ciA, ciB, ciF: Confidence intervals for the test statistics.

-TsA, TsB, TsF: Bootstrap test statistics.

-modelo: The linear model object.

-ihto: The initial iht computation result.

References

Dwivedi, A. K., Mallawaarachchi, I., & Alvarado, L. A. (2017). Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method. Statistics in Medicine, 36(14), 2187–2205. https://doi.org/10.1002/sim.7263

Vanbrabant, L., & Rosseel, Y. (2020). An Introduction to Restriktor: Evaluating informative hypotheses for linear models. In R. van de Schoot & M. Miocevic (Eds.), Small Sample Size Solutions: A Guide for Applied Researchers and Practitioners (1st ed., pp. 157 -172). Routledge. https://doi.org/10.4324/9780429273872-14

Examples

bootFbar(data = iris, formula = Sepal.Length ~ -1 + Species,
         grp = "Species",
         constraints = 'Speciessetosa < Speciesversicolor < Speciesvirginica',
         nboot = 10, conf.level = 0.95, seed = NULL, na_rm = FALSE)

---

Performance Simulation On Type I Error or Statistical Power

Description

Simulates performance (Type I Error or Statistical Power) for a range of statistical tests, including bootFbar, default setting restriktor::iht, and parametric bootstrap restriktor::iht. The function is set up to replicate the table structure in Dwivedi et al. (2017) Supplemental Tables 2 & 3.

Usage

persimon(
  M1 = 5,
  S1 = 1,
  M2 = 5,
  S2 = 1,
  M3 = 5,
  S3 = 1,
  Sk1 = NULL,
  Sk2 = NULL,
  Sk3 = NULL,
  n1 = c(2, 3, 4, 5, 6, 7, 8, 9, 10, 15),
  n2 = c(2, 3, 4, 5, 6, 7, 8, 9, 10, 15),
  n3 = c(2, 3, 4, 5, 6, 7, 8, 9, 10, 15),
  n_simulations = 10000,
  nboot = 1000,
  conf.level = 0.95
)

Arguments

M1	Mean value for group 1.
S1	Standard deviation for group 1.
M2	Mean value for group 2.
S2	Standard deviation for group 2.
M3	Mean value for group 3.
S3	Standard deviation for group 3.
Sk1	Skewness parameter for group 1; NULL for normal distribution.
Sk2	Skewness parameter for group 2; NULL for normal distribution.
Sk3	Skewness parameter for group 3; NULL for normal distribution.
n1	Vector of sample sizes for group 1.
n2	Vector of sample sizes for group 2.
n3	Vector of sample sizes for group 3.
n_simulations	Number of simulations to run (default is 10000).
nboot	Number of bootstrap samples (default is 1000).
conf.level	Confidence level for the tests (default is 0.95).

Details

The persimon function generates data for three groups with specified mean, standard deviation, and skewness, and then applies a range of statistical tests to this data, simulating the process across a specified number of iterations to assess performance in terms of Type I Error or Statistical Power. The output mirrors the format of the supplemental tables in Dwivedi et al. (2017).

Value

A list of data frames. One with the proportions of rejecting the null hypothesis for each test and sample size combination, and the other with the number of models which did not produce errors for each combination.

References

Dwivedi, A. K., Mallawaarachchi, I., & Alvarado, L. A. (2017). Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method. Statistics in Medicine, 36(14), 2187–2205. https://doi.org/10.1002/sim.7263

Examples

set.seed(135)
persimon(M1 = 5, S1 = 1, M2 = 5, S2 = 1, M3 = 5, S3 = 1,
         n1 = 6, n2 = 6, n3 = 6, n_simulations = 2, nboot = 4,
         conf.level = 0.95)

---

Run Shiny Application for bootFbar Performance Simulation

Description

This function launches a 'shiny' application that allows users to run performance simulations for the bootFbar method and other statistical tests. The app provides an interface to set simulation parameters, run simulations, view results, and store them in a 'PostgreSQL' database.

Usage

persimon_app(dbname, datatable, host, port, user, password)

Arguments

dbname	A string specifying the name of the 'PostgreSQL' database to connect to.
datatable	A string specifying the name of the table in the database where results will be stored.
host	A string specifying the host name or IP address of the 'PostgreSQL' server.
port	An integer specifying the port number on which the 'PostgreSQL' server is listening.
user	A string specifying the username for the 'PostgreSQL' database connection.
password	A string specifying the password for the 'PostgreSQL' database connection.

Details

The 'shiny' application provides a user interface for setting simulation parameters, running simulations using the persimon function, and visualizing the results. It allows users to:

• Set parameters for the simulation (means, standard deviations, sample sizes, etc.)

• Run simulations with the specified parameters

• View simulation results and success rates in interactive tables

• Store simulation results in a PostgreSQL database

• Download stored results from the database

• View relevant citations for the methods used

The application uses the mmints package for database interactions and citation handling.

Value

A 'shiny' application object.

See Also

persimon for the underlying simulation function.

Examples

if(interactive()){
persimon_app(dbname = "my_database", datatable = "simulation_results",
             host = "localhost", port = 5432,
             user = "username", password = "password")
}

---