*Accelerate Bayesian analytics workflows* in R through interactive
modelling, visualization, and inference. Uses probabilistic graphical
models as a unifying language for business stakeholders, statisticians,
and programmers.

This package relies on the sleek and elegant `greta`

package for
Bayesian inference. `greta`

, in turn, is an interface into `TensorFlow`

from R. Future iterations of the `causact`

package will aim to be a
front-end into several universal probablistic programming languages
(e.g. Stan, Turing, Gen, etc.).

Using the `causact`

package for Bayesian inference is featured in ```
A
Business Analyst's Introduction to Business Analytics
```

available at
https://www.causact.com/.

NOTE: Package is under active development. Breaking changes are to be expected. Feedback and encouragement is appreciated via github issues or Twitter (https://twitter.com/preposterior).

You can install the current release version of the package from CRAN:

```
install.packages("causact")
```

or the development version from GitHub:

```
install.packages("remotes")
remotes::install_github("flyaflya/causact")
```

`causact`

requires the `greta`

package for Bayesian updating, which in
turn, requires a specific version of `TensorFlow`

. Install both `greta`

and `TensorFlow`

using the instructions available here:
https://www.causact.com/install-tensorflow-greta-and-causact.html.

Example taken from
https://www.causact.com/graphical-models-tell-joint-distribution-stories.html#graphical-models-tell-joint-distribution-stories
with the packages `dag_foo()`

functions further described here:

```
library(causact)
graph = dag_create() %>%
dag_node(descr = "Get Card", label = "y",
rhs = bernoulli(theta),
data = carModelDF$getCard) %>%
dag_node(descr = "Card Probability", label = "theta",
rhs = beta(2,2),
child = "y") %>%
dag_plate(descr = "Car Model", label = "x",
data = carModelDF$carModel,
nodeLabels = "theta",
addDataNode = TRUE)
graph %>% dag_render()
```

```
graph %>% dag_render(shortLabel = TRUE)
```

`greta`

code without executing it (for debugging or learning)```
library(greta)
#>
#> Attaching package: 'greta'
#> The following objects are masked from 'package:stats':
#>
#> binomial, cov2cor, poisson
#> The following objects are masked from 'package:base':
#>
#> %*%, apply, backsolve, beta, chol2inv, colMeans, colSums, diag,
#> eigen, forwardsolve, gamma, identity, rowMeans, rowSums, sweep,
#> tapply
gretaCode = graph %>% dag_greta(mcmc = FALSE)
#> ## The below greta code will return a posterior distribution
#> ## for the given DAG. Either copy and paste this code to use greta
#> ## directly, evaluate the output object using 'eval', or
#> ## or (preferably) use dag_greta(mcmc=TRUE) to return a data frame of
#> ## the posterior distribution:
#> y <- as_data(carModelDF$getCard) #DATA
#> x <- as.factor(carModelDF$carModel) #DIM
#> x_dim <- length(unique(x)) #DIM
#> theta <- beta(shape1 = 2, shape2 = 2, dim = x_dim) #PRIOR
#> distribution(y) <- bernoulli(prob = theta[x]) #LIKELIHOOD
#> gretaModel <- model(theta) #MODEL
#> meaningfulLabels(graph)
#> draws <- mcmc(gretaModel) #POSTERIOR
#> drawsDF <- replaceLabels(draws) %>% as.matrix() %>%
#> dplyr::as_tibble() #POSTERIOR
#> tidyDrawsDF <- drawsDF %>% addPriorGroups() #POSTERIOR
```

`greta`

code```
library(greta)
drawsDF = graph %>% dag_greta()
drawsDF ### see top of data frame
#> # A tibble: 4,000 x 4
#> theta_JpWrnglr theta_KiaForte theta_SbrOtbck theta_ToytCrll
#> <dbl> <dbl> <dbl> <dbl>
#> 1 0.854 0.195 0.597 0.191
#> 2 0.824 0.243 0.628 0.195
#> 3 0.824 0.243 0.628 0.195
#> 4 0.855 0.264 0.593 0.202
#> 5 0.829 0.219 0.594 0.189
#> 6 0.835 0.262 0.588 0.185
#> 7 0.857 0.241 0.613 0.210
#> 8 0.864 0.246 0.610 0.208
#> 9 0.821 0.229 0.529 0.219
#> 10 0.834 0.243 0.571 0.239
#> # ... with 3,990 more rows
```

```
drawsDF %>% dagp_plot()
```

Credible interval plots.

For more info, see ```
A Business Analyst's Introduction to Business
Analytics
```

available at https://www.causact.com. Two additional
examples are shown below.

McElreath, Richard. Statistical rethinking: A Bayesian course with examples in R and Stan. Chapman and Hall/CRC, 2018.

```
library(greta)
library(tidyverse)
library(causact)
# data object used below, chimpanzeesDF, is built-in to causact package
graph = dag_create() %>%
dag_node("Pull Left Handle","L",
rhs = bernoulli(p),
data = causact::chimpanzeesDF$pulled_left) %>%
dag_node("Probability of Pull", "p",
rhs = ilogit(alpha + gamma + beta),
child = "L") %>%
dag_node("Actor Intercept","alpha",
rhs = normal(alphaBar, sigma_alpha),
child = "p") %>%
dag_node("Block Intercept","gamma",
rhs = normal(0,sigma_gamma),
child = "p") %>%
dag_node("Treatment Intercept","beta",
rhs = normal(0,0.5),
child = "p") %>%
dag_node("Actor Population Intercept","alphaBar",
rhs = normal(0,1.5),
child = "alpha") %>%
dag_node("Actor Variation","sigma_alpha",
rhs = exponential(1),
child = "alpha") %>%
dag_node("Block Variation","sigma_gamma",
rhs = exponential(1),
child = "gamma") %>%
dag_plate("Observation","i",
nodeLabels = c("L","p")) %>%
dag_plate("Actor","act",
nodeLabels = c("alpha"),
data = chimpanzeesDF$actor,
addDataNode = TRUE) %>%
dag_plate("Block","blk",
nodeLabels = c("gamma"),
data = chimpanzeesDF$block,
addDataNode = TRUE) %>%
dag_plate("Treatment","trtmt",
nodeLabels = c("beta"),
data = chimpanzeesDF$treatment,
addDataNode = TRUE)
```

```
graph %>% dag_render(width = 2000, height = 800)
```

```
graph %>% dag_render(shortLabel = TRUE)
```

```
drawsDF = graph %>% dag_greta()
```

```
drawsDF %>% dagp_plot()
```

Gelman, Andrew, Hal S. Stern, John B. Carlin, David B. Dunson, Aki Vehtari, and Donald B. Rubin. Bayesian data analysis. Chapman and Hall/CRC, 2013.

```
library(greta)
library(tidyverse)
library(causact)
# data object used below, schoolDF, is built-in to causact package
graph = dag_create() %>%
dag_node("Treatment Effect","y",
rhs = normal(theta, sigma),
data = causact::schoolsDF$y) %>%
dag_node("Std Error of Effect Estimates","sigma",
data = causact::schoolsDF$sigma,
child = "y") %>%
dag_node("Exp. Treatment Effect","theta",
child = "y",
rhs = avgEffect + schoolEffect) %>%
dag_node("Pop Treatment Effect","avgEffect",
child = "theta",
rhs = normal(0,30)) %>%
dag_node("School Level Effects","schoolEffect",
rhs = normal(0,30),
child = "theta") %>%
dag_plate("Observation","i",nodeLabels = c("sigma","y","theta")) %>%
dag_plate("School Name","school",
nodeLabels = "schoolEffect",
data = causact::schoolsDF$schoolName,
addDataNode = TRUE)
```

```
graph %>% dag_render()
```

```
drawsDF = graph %>% dag_greta()
```

```
drawsDF %>% dagp_plot()
```

```
#### use dirichlet instead
library(greta)
library(tidyverse)
library(causact)
## sample data - try to recover params
x <- c(rpois(800, 3),rpois(200, 10))
graph = dag_create() %>% ## create generative DAG
dag_node("Mixed Var","x",
rhs = mixture(alpha,beta,
weights = t(weights)),
data = x) %>%
dag_node("Count Var 1","alpha",
rhs = poisson(lambda1),
child = "x") %>%
dag_node("Count Var 2","beta",
rhs = poisson(lambda2),
child = "x") %>%
dag_node("Weight Vars","weights",
rhs = dirichlet(t(c(1,1))),
child = "x") %>%
dag_node("Exp Rate 1","lambda1",
rhs = uniform(1,5),
child = "alpha") %>%
dag_node("Exp Rate 2","lambda2",
rhs = uniform(6,20),
child = "beta")
```

```
graph %>% dag_render()
```

```
drawsDF %>% dagp_plot()
```

**Any scripts or data that you put into this service are public.**

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.