1 R: setup and resources

This chapter is for getting R itself ready and, if you need it, for brushing up on the language. Everything energyRt-specific — the package, the solver backends — is in the next chapter, Installation.

You do not need to be an experienced R user to follow the series. Familiarity helps, but the sessions explain the modeling ideas as they go, and you are welcome to attend simply to watch.

1.1 Install R and an IDE

What	Where	Needed?
R (≥ 4.3)	https://cloud.r-project.org/	Required
RStudio Desktop — the IDE we use in the sessions	https://posit.co/downloads	Strongly recommended
Quarto — publishing system for reports and slides	https://quarto.org/	Optional

Install R first, then RStudio: RStudio is a front end and looks for an existing R installation when it starts.

Recent versions of RStudio bundle Quarto, so a separate install is usually unnecessary — it is listed here because Quarto is worth knowing about in its own right, and the course materials are written in it.

Windows: you also want Rtools

energyRt is installed from source, which needs a compiler toolchain. On Windows that is Rtools — install the version matching your R (Rtools 4.5 for R 4.5.x). A useful side effect: Rtools ships the GLPK solver we use in the training, so installing it covers two requirements at once. See Installation.

1.2 Learning (or refreshing) R

Many good introductions exist; these are a few short ones that get you to a working knowledge quickly. Pick one — they cover much the same ground.

Introduction to R — a gentle, complete beginner’s book: installing, data structures, plotting, basic programming.
R intro (Otoomet, INFO 201) — compact and fast-moving; good if you program already and just want R’s idioms.
R for Data Science (2e) — the standard reference for the tidyverse workflow that energyRt’s data handling follows.
Hands-On Programming with R — focuses on the language itself (objects, functions, environments) rather than data analysis.
swirl — learn interactively inside the R console, if you prefer doing to reading.

Time is short before the first session — the highest-return preparation is being comfortable with data frames, |> pipes, and subsetting. Everything else you can pick up as we go.

1.3 Key packages to check first

Install these before Session 1 and confirm they load. They are the ones most likely to need a compiler or a system library, so finding a problem now rather than mid-session is worth the few minutes:

install.packages(c("dplyr", "ggplot2", "sf"))

library(dplyr)     # data manipulation — the verbs energyRt returns results for
library(ggplot2)   # plots
library(sf)        # spatial data and maps

dplyr — results come back from energyRt as tidy tables; filter(), group_by(), summarise() are how you work with them.
ggplot2 — used throughout for charts of capacity, dispatch, and costs.
sf — spatial features, for models with regions and maps. It is the one most likely to complain, since it links to the GDAL/GEOS/PROJ system libraries. On Windows and macOS the CRAN binary includes them; on Linux install libgdal-dev libgeos-dev libproj-dev (or your distribution’s equivalents) first.

If library(sf) fails, it will not block you from the modeling parts of the course — only from the mapping examples.

1.4 Cheatsheets

Posit maintains one-page reference cards for most of the tools we use. Printing or bookmarking a few is the single most useful thing you can do before the sessions:

https://opensource.posit.co/resources/cheatsheets/

Worth having open during the course:

Data transformation with dplyr — the verbs, joins, and grouping
Data visualization with ggplot2 — geoms, scales, facets
RStudio IDE — panes, shortcuts, projects
R Markdown and Quarto — for writing up your own results
sf / Simple Features — if you get to the spatial examples

1.5 Getting help

?function_name in the console opens R’s own documentation — always the fastest first stop.
Bring anything unresolved to the course Q&A thread — it is there precisely for this, and beginner questions are welcome.