---
title: "Getting Started with buzzMed"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Getting Started with buzzMed}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```
## Introduction

`buzzMed` is a collection of quantitative tools for selecting mediating effects within exploratory Bayesian mediation models. The package accommodates both continuous and dichotomous outcomes, including the dependent variables and the mediators for identifying and analyzing mediation pathways.

## Basic Usage

First, load the package:

```{r setup}
library(buzzMed)
```

To run a basic mediation model, use one of the `buzzEBM` functions depending on
your data types (continuous or categorical):

```{r example, eval = FALSE}
library(buzzMed)

# Create some toy data to play with
my_data <- data.frame(
    MyPredictor = rnorm(30),
    MyMediator1 = rnorm(30),
    MyMediator2 = rnorm(30),
    MyOutcome = rnorm(30)
)

# Specify your mediation model using syntax 'Y ~ M1 + M2 | X'
model_string <- "MyOutcome ~ MyMediator1 + MyMediator2 | MyPredictor"

# Run the model with continuous mediator and continuous outcome
fit <- buzzEBMcontMcontY(model = model_string, dataset = my_data)
```

## More Information

For detailed theory on the Bayesian mediation selection process, please refer to
the package documentation for the specific `buzzEBM` functions.

## References
> Shi, D., Dexin Shi, & Amanda J. Fairchild (2023). Variable Selection for Mediators under a Bayesian Mediation Model. Structural Equation Modeling: A Multidisciplinary Journal, 30(6), 887-900. DOI: 10.1080/10705511.2022.2164285