R · Quarto setup

Getting set up to run the simulation code

The inference labs use R and Quarto to simulate sampling distributions, draw likelihood and posterior curves, build bootstrap intervals, and run randomization tests. This page gets you set up and explains the reproducibility conventions every lab uses. You need no prior R experience — the simulation work is scaffolded in the labs, and the code is explained as it goes.

Note

On this site the R code in the notes and labs is shown as static teaching code — it is displayed, syntax-highlighted, but not executed in place. You run it yourself in your own R session, which is exactly how you will work on the labs.

Option A — Posit Cloud (nothing to install)

The quickest start is the browser. Create a free account at posit.cloud, open a new project, and you have R, RStudio, and Quarto ready to use with nothing installed on your machine. This is the recommended path if you are new to R or working on a shared computer.

Option B — install locally

To work offline, install two pieces:

  1. R — the language — from cran.r-project.org.
  2. RStudio Desktop — the editor — from posit.co/download/rstudio-desktop. Recent RStudio ships with Quarto built in; if you want the standalone tool, get it from quarto.org.

Open RStudio, create a new Quarto document (File → New File → Quarto Document), and you are ready.

Running the lab code

Each lab is a sequence of small R chunks. Copy a chunk into your Quarto document (or an R script), run it, and read the output against the lab’s Verify section. The labs use only base R — rbinom, rnorm, replicate, sample, quantile, dbinom, qbeta, pbeta, and optimize — so there is nothing extra to install for the simulations.

Reproducibility conventions

Three habits make your work reproducible, and the labs use all three:

  • Fix the seed. Begin with set.seed(35103) so every simulation returns the same numbers on every run. This is what lets your output match the locked values in the notes — and what lets a reader confirm your results.
  • One Quarto file, top to bottom. Keep the narrative, the code, and the output in a single .qmd that renders start to finish without manual steps.
  • Record session information. End a report with sessionInfo() so the exact R and package versions are part of the record.
set.seed(35103)   # put this near the top, before any random draw
# ... your simulation code ...
sessionInfo()     # at the end, to capture versions

If something goes wrong

  • Numbers change every run. You did not set the seed, or set it after a random draw. Move set.seed(35103) above the simulation.
  • A jittery histogram or interval. Your number of repetitions is too small; push replicate(...) or the bootstrap count to \(10{,}000\).
  • could not find function. A typo in a base-R function name, or you are in a fresh session — re-run the setup chunk.

A note on AI help

You may use an AI assistant to explain a function or help debug your own code, but you must check what it produces — re-run it with the fixed seed, compare to the lab’s Verify section, and include an AI Use Note (Tool / Purpose / Verification) on any graded work. Verification is the load-bearing line.

This page is a study reference. For graded specifics — deadlines, submissions, and policies — Blackboard (the LMS) is authoritative.