---
title: "Interactive Data Analysis with MAIHDA"
author: "Hamid Bulut"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Interactive Data Analysis with MAIHDA}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

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

## Introduction

The **MAIHDA** package includes a fully-featured interactive Shiny Dashboard that provides a no-code alternative to exploring your data, building intersectional strata, fitting models, and analyzing inequality. This is particularly useful for rapid exploration.

## Launching the Application

### Online Version

You can access a live, cloud-hosted version of the MAIHDA interactive dashboard directly in your browser without installing R:
**[https://hdbt.shinyapps.io/shiny/](https://hdbt.shinyapps.io/shiny/)**

### Local Version

You can start the interactive dashboard locally by running a single command:

```{r eval=FALSE}
library(MAIHDA)
run_maihda_app()
```

This will automatically launch the dashboard in your default web browser or the RStudio viewer.

```{r, echo=FALSE, out.width="100%", fig.cap="Shiny Dashboard", eval=TRUE}
 knitr::include_graphics("figures/app_data_import.png")
```


## App Features

### 1. Data Import

* **Upload Own Data:** Easily upload datasets in `.csv`, Stata (`.dta`), or SPSS (`.sav`) formats.
* **Use Included Data:** Try out the app instantly by selecting the pre-loaded built-in **NHANES health data** (`maihda_health_data`) or **PISA country data** (`maihda_country_data`).
* **View Data:** The app includes an interactive data table letting you sort, filter, and inspect variables before analyzing.


### 2. Variable Selection & Strata Creation

* Choose a categorical/continuous outcome metric from your dataset.
* Select two or more categorical demographic variables (e.g., gender, race, education) to automatically generate intersectional strata.

### 3. Model Fitting & Settings

* Fit models with the **lme4** engine used by the interactive dashboard. Bayesian **brms** models remain available from R code via `fit_maihda(engine = "brms")`.
* Select covariates to control for within your models.
* Choose whether to calculate **parametric bootstrap confidence intervals** to quantify uncertainty. When enabled, the dashboard reports bootstrap intervals for both the Variance Partition Coefficient (VPC / ICC) -- shown in the Model Summary and Interactive Explorer tabs -- and the Proportional Change in Variance (PCV), shown in the PCV Results tab.

### 4. Interactive Visualizations

Once a model is fit, you can navigate across multiple tabs:

* **Predicted Values:** Visually evaluate stratum-level predictions relative to the overall mean with dynamic prediction intervals.
* **VPC Decomposition:** Examine how much of your outcome's variance is attributed to between-stratum differences versus within-stratum individual heterogeneity.
* **Observed vs. Shrunken Estimates:** Compare raw unadjusted group means to your model's shrinkage estimates to see the protective mechanism of multilevel modeling.

### 5. Stepwise Variance Analysis (PCV)

The dashboard calculates stepwise Proportional Change in Variance (**PCV**) tables:

* See the proportional change in between-stratum variance as covariates are added sequentially (a model-dependent, descriptive comparison, not a causal decomposition of inequality).
* Uncover masking/suppression effects directly inside the app by comparing partial PCV values across models.

## Summary

The MAIHDA interactive dashboard is designed to make modeling health and social inequalities accessible without needing to write code. It provides a platform for exploring intersectional data, fitting multilevel models, and visualizing results in a user-friendly way.