Demo repository containing stock portfolio assets (Shiny, RMarkdown, Plumber, ...)
These demos showcase different components of RStudio Connect. To view the final results, visit: http://colorado.rstudio.com/rsc and filter by the tags: Finance > Stocks > Portfolio.
Each folder contains a piece of R content. More details on each are below. There are
also two SQLite databases with sample data (./stock.db and ./portfolio.db) to make the repository and resources somewhat encapsulated from external dependencies.
-
The
shiny-appfolder has a R Markdown flexdashboard that uses shiny to allow users to explore portfolio returns based on different tolerances for risk. -
The
apifolder surfaces a function that computes the volatiltiy of a stock price as a RESTful API. RStudio Connect automatically scales the number of R processes based on the volume of incoming API requests. RStudio Connect also contains a record of deployments, making it easy to roll back to a prior version of the API. -
The
report-portfoliofolder contains a parameterized R Markdown report that can be used to do batch exploration of the same portfolio returns explored in the Shiny application above. The report generates a customized email for remote review by stakeholders and RStudio Connect stores the report history for review over time. -
The
report-stockfolder shows a quick summary of a specific stock ticker symbol. The R Markdown report is parameterized, making it easy to schedule different versions of the report in RStudio Connect. The report also generates an email that includes a custom subject line, embeds a plot and table in the body of the email, and attaches a summary of the data in a csv file. RStudio Connect stores the history of the report and makes it easy to compare versions of the report over time. -
The
etlfolder extends the use of RStudio Connect's job scheduler to an ETL task that updates a database. The ETL document provides a quick visualization to validate the model over time and RStudio Connect sends an email if the ETL task fails.