Dynamic Presentations with Stata and Markdown

Germán Rodríguez, Princeton University

7 November 2017

Bullet Points

You presentation will usually include bullet points:

These can be shown incrementally.

The same script can generate an HTML presentation using the S5 engine, or PDF slides via LaTeX using Beamer.

Stata Output

Here’s the proverbial fuel efficiency dataset:

. sysuse auto, clear
(1978 Automobile Data)

. gen gphm = 100/mpg

. quietly reg gphm weight

. _coef_table
        gphm │      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
      weight │    .001407   .0001008    13.95   0.000      .001206    .0016081
       _cons │   .7707669   .3142571     2.45   0.017     .1443069    1.397227

So a car that weights 1,000 pounds more than another will need on average 1.4 more gallons to travel 100 miles.

Stata Graphs

Using gallons per 100 miles leads to a more linear relationship

. twoway scatter gphm weight || lfit gphm weight, ///
>     legend(off) ytitle(Gallons per 100 miles)

. quietly graph export gphm.png, width(800) replace
Fuel Efficiency by Weight

Fuel Efficiency by Weight

Notes on Features

This example uses the simple syntax, with Stata code indented one tab or four spaces.

If you get carried away with nested lists, or wish to hide code, you can use the strict syntax.

As usual, you can include math formulas and/or bibliographic references.

The fragile attribute is used to keep Beamer happy with verbatim blocks, and is ignored in HTML

A nodo option let’s you tweak the narrative or change engines and/or themes without having to rerun the Stata code, an idea from Ben Jann’s texdoc.