Section - 1 Setup R Environment

1.2 Replicate Results

The paper will have to be a static file because that’s how a research paper would work obviously, so there will be two main ways to follow along with the tutorial:

1.2.1 Reproducible Example

Reproducible examples are extremely important to have in the research community. The costs associated with this issue are quite massive, so I think the paper could/should have a section around creating reproducible examples and how to do that in R could be incredibly valuable. There are several amazing packages that were created by the team at RStudio for this specific purpose, most notably the reprex package by Jenny Bryan. That package along with a brand new package called pins make creating reproducible examples in R incredibly easy, so I think it makes sense to add it as its own section.

Regardless of whether this will be its own section with a tutorial around creating them or not, one of the ways to follow along will be to use the exact same data that we use when we publish it. So if we used 14 months worth of data and created the whole paper using that data, anyone could use the exact same dataset in a completely identical way and get the exact same results that we publish.

1.2.2 Latest Data

The second option when following along with the tutorial, will be to re-create the same type of analysis, but using the latest data. In this case, the reader would follow along our research that uses the entire dataset at the time of publication, but as they follow those same steps they would only analyze the last 7 days (or 31 days, TBD) worth of data, and create a brand new analysis that is as current as possible.

My versions will do this to keep the runtimes faster, but the paper itself including the reproducible example will have as complete a dataset as reasonably possible.

## [1] TRUE
## Warning: Closing open result sets
## [1] TRUE

References

Wickham, Hadley. 2019. Tidyverse: Easily Install and Load the ’Tidyverse’. https://CRAN.R-project.org/package=tidyverse.