### Install rstanarm Development Version from GitHub

Source: https://github.com/stan-dev/rstanarm/blob/master/README.md

Installs the latest development version of rstanarm from GitHub. Ensure rstan and a C++ toolchain are installed first. Adjust '-j2' based on available CPU cores.

```r
# Change 2 to however many cores you can/want to use to parallelize install
# If you experience crashes or run out RAM during installation, try changing this to 1
Sys.setenv(MAKEFLAGS = "-j2")
Sys.setenv("R_REMOTES_NO_ERRORS_FROM_WARNINGS" = "true")
remotes::install_github("stan-dev/rstanarm", INSTALL_opts = "--no-multiarch", force = TRUE)
```

--------------------------------

### Example Model Fitting with stan_glmer

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.md

Demonstrates fitting a generalized linear mixed-effects model using `stan_glmer` from the rstanarm package. This example is kept small for demonstration purposes by limiting chains and iterations.

```R
### script to run all rstanarm tests locally

library(rstanarm)
# library(rstantools)
# library(bayesplot)

remove(list=ls())
### run prerequisite functions
#
example_model <-
  stan_glmer(cbind(incidence, size - incidence) ~ size
  + period + (1|herd),
             data = lme4::cbpp, family = binomial,
             # this next line is only to keep the example small in size!
             chains = 2, cores = 1, seed = 12345, iter = 500)
#
last_dimnames <- function(x) {
  ndim <- length(dim(x))
  dimnames(x)[[ndim]]
}

### run tests
devtools::test()
```

--------------------------------

### Install rstanarm from CRAN

Source: https://github.com/stan-dev/rstanarm/blob/master/README.md

Installs the latest stable release of the rstanarm package from the Comprehensive R Archive Network (CRAN).

```r
install.packages("rstanarm")
```

--------------------------------

### Install Survival Analysis Version of rstanarm

Source: https://github.com/stan-dev/rstanarm/blob/master/README.md

Installs a binary version of the survival branch of rstanarm from the Stan R packages repository. This is useful for accessing experimental survival analysis functionality without compiling from source.

```r
install.packages("rstanarm", repos = c('https://stan-dev.r-universe.dev', getOption("repos")))
```

--------------------------------

### List S3 Methods for 'ivreg' Class

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

This command lists the available S3 methods for objects of class 'ivreg'. It helps identify which post-estimation functions might need to be reimplemented for compatibility with rstanarm.

```r
methods(class = "ivreg")
```

--------------------------------

### RStanArm Model Fit Workflow

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.md

Illustrates the typical workflow for fitting models using rstanarm, showing the interaction between user-defined R functions, the .fit workhorse function, and the Stan model file.

```text
                     ┏━━━━━━ USER ━━━━━━┓
                  model_A.R                  model_B.R
                      ┗━━━━ model.fit ━━━━┛
                                   ┃
                               model.stan
```

--------------------------------

### Run rstanarm Tests Locally

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.html

A script to execute all rstanarm tests locally. It loads the rstanarm library and runs prerequisite functions before initiating the tests.

```R
library(rstanarm)
# library(rstantools)
# library(bayesplot)

remove(list=ls())
### run prerequisite functions
#
example_model <-
  stan_glmer(cbind(incidence, size - incidence) ~ size
      + period + (1|herd),
                 data = lme4::cbpp, family = binomial,
                 # this next line is only to keep the example small in size!
                 chains = 2, cores = 1, seed = 12345, iter = 500)
#
last_dimnames <- function(x) {
  ndim <- length(dim(x))
  dimnames(x)[[ndim]]
}

### run tests
devtools::test()
```

--------------------------------

### Create Git Branch for New Model

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

Use git to create a new branch for your development work. The branch name should reflect the functionality being added.

```git
git checkout master
git checkout -b feature/ivreg
```

--------------------------------

### Inspect Function Arguments with args()

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

Use the args() function to inspect the arguments of an existing R function, which helps in emulating its signature in a new rstanarm function.

```R
> args(AER::ivreg)
function (formula, instruments, data, subset, na.action, weights, 
    offset, contrasts = NULL, model = TRUE, y = TRUE, x = FALSE, 
    ...)
```

--------------------------------

### Include Vignette Children Snippets

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.html

Demonstrates how to include reusable code snippets from other files into vignettes using the child directive.

```R
{r, child="children/*.txt"}
```

--------------------------------

### Emulate AER::ivreg Data Parsing in R

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

Use this code to parse data in the same way as AER::ivreg. It constructs a call to AER::ivreg, evaluates it, and captures the resulting model frame.

```r
mc <- match.call(expand.dots = FALSE)
mc$model <- mc$y <- mc$x <- TRUE
# NULLify any Stan specific arguments in mc now
mc[[1L]] <- quote(AER::ivreg)
if (!requireNamespace("AER")) stop("the AER package is needed by 'stan_ivreg'")
mf <- eval(mc, parent.frame())
```

--------------------------------

### Emulate R Function Arguments in stan_foo

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

Define the stan_foo function by adopting arguments from an existing R function. The '...' argument is used to separate arguments from the emulated function and those added for rstanarm, and to pass additional arguments to the underlying sampling function.

```R
stan_ivreg <- function (formula, instruments, data, subset, na.action, weights, 
                        offset, contrasts = NULL, model = TRUE, y = TRUE, x = FALSE, 
                        ...) # there will be more arguments
```

--------------------------------

### Set knitr Chunk Options

Source: https://github.com/stan-dev/rstanarm/blob/master/vignettes/children/SETTINGS-knitr.txt

Configures default knitr chunk options for RStanarm documentation. This includes suppressing messages and warnings, controlling evaluation based on the NOT_CRAN environment variable, and setting default figure properties like device, resolution, aspect ratio, width, output width, and alignment.

```R
stopifnot(require(knitr))
opts_chunk$set(
  comment=NA, 
  message = FALSE, 
  warning = FALSE, 
  eval = identical(Sys.getenv("NOT_CRAN"), "true"),
  dev = "png",
  dpi = 150,
  fig.asp = 0.618,
  fig.width = 5,
  out.width = "60%",
  fig.align = "center"
)
```

--------------------------------

### Set Default ggplot2 Theme

Source: https://github.com/stan-dev/rstanarm/blob/master/vignettes/children/SETTINGS-gg.txt

Loads necessary libraries and sets the default ggplot2 theme to a bayesplot theme. This is useful for ensuring consistent plot styles across multiple visualizations.

```r
library(ggplot2)
library(bayesplot)
theme_set(bayesplot::theme_default())
```

--------------------------------

### View Compiled Stan Model in R

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.md

Access and view the compiled Stan model object named 'continuous' within the rstanarm package.

```R
rstanarm:::stanmodels$continuous
```

--------------------------------

### Stan Generated Quantities for Intercept Transformation

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.md

Shows how to transform the intercept parameter in the generated quantities block when covariates have been centered. This is necessary to report the correct intercept value.

```stan
generated quantities {
  array[has_intercept] real alpha;
  if (has_intercept) {
    alpha[1] = gamma[1] - dot_product(beta, xbar)
  }
}
```

--------------------------------

### Set Normal Prior for Intercept and Coefficients

Source: https://github.com/stan-dev/rstanarm/blob/master/vignettes/children/stan_glm_priors.txt

Use `normal(0,1)` for both `prior_intercept` and `prior` to specify a normal distribution with mean 0 and scale 1 for the intercept and all regression coefficients. This represents a belief that parameters are equally likely to be positive or negative but unlikely to be far from zero.

```R
stan_glm(formula, data = NULL, prior_intercept = normal(0, 1), prior = normal(0, 1))
```

--------------------------------

### Define stan_ivreg.fit function in rstanarm

Source: https://github.com/stan-dev/rstanarm/wiki/Adding-a-new-model-to-rstanarm

This function serves as the low-level interface for fitting instrumental variable regression models. It includes argument validation to ensure data integrity before passing to the Stan fitting function.

```R
stan_ivreg.fit <- function (x, y, z, weights, offset, ...) { # there will be more arguments
  stopifnot(is.matrix(X), is.vector(y), is.matrix(z), length(y) == nrow(X), nrow(X) == nrow(Z))
}
```

--------------------------------

### Stan Intercept Transformation

Source: https://github.com/stan-dev/rstanarm/blob/master/dev-notes/rstanarm_dev_notes.html

Transform the intercept parameter when using centered predictors in Stan. This is necessary to correctly report the intercept after fitting with centered covariates.

```stan
generated quantities {
  real alpha[has_intercept];
  {
    alpha[1] = gamma[1] - dot_product(beta, xbar)
  }
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.