This repository contains research into the characteristics of statistical indicators used for ESG sovereign wealth analysis. It is an RMarkdown project that uses the bookdown project to produce a paper available in either HTML or PDF formats. Most of the analytical code is written in R, but some functions are written in Python and integrated with the reticulate package. The outputs are available at https://worldbank.github.io/ESG_gaps_research
We built with:
In RStudio, choose "bookdown::tufte_book2" from the "Build Book" menu in the book tab
In RStudio, choose "bookdown::tufte_html_book" from the "Build Book" menu in the book tab
There is an apparent bug in the HTML system that is preventing MathJax
equations (used in Appendix 2) from rendering correctly, and requires a post-build
fix. The apparent culprit is plotly-latest.min.js so the easiest fix is to remove
this reference since it isn't needed in the appendix.
In RStudio, you can run this code from the console after the build:
source("R/report.R")
mathjax_fix()Then commit changes produced by the build and push them to github.
One significant challenge of this project was designing visual content that worked in
both PDF and HTML outputs. Several additions were necessary on the PDF/LaTeX side
(see mystyles.sty). Interactive tables and graphs in HTML needed to be converted
to static outputs in PDF. This was accomplished through a set of wrapper functions
that provided reasonable defaults for each output while allowing output-specific
customization. These functions may all be found in R/report.R and are designed to
be resuable in similar projects with perhaps some changes to the default options.