Ten years after — v2.0

[briatte]

This course is now roughly 10 13 years old (first run: Fall 2010). The pretty dirty repo history shows it. It's time to think of version 2.0, although version 1.0 never got its release tag (release tags did not even exist when we started).

• Update the overall code to whatever Stata compatibility we choose:
  • Compatibility with different versions of Stata #28
  • Test with version n
  • Add version 13 to 'freeze' some commands, e.g. table, margins?
  • Test datasets with Stata < 13
• Update the datasets
  • Country-level data: World Economics and Politics Dataverse #24
  • Many QOG variables have low-N for cross-sectional analysis #29
  • Dataset updates #30 (includes 'max vars' issue from Compatibility with different versions of Stata #28)
• Update the utilities in setup, especially unpublished ones; see also svyplot #26
• List all instructors and admin: Try to list all former instructors (and admin) #36
• Rebase the repository for cleanliness?
• Dig up the very old stuff, for fun: lobbying.dta (Baumgartner), ebm2009 (Eurobarometer)

And perhaps even more importantly, but (perhaps, even) more time-consumingly:

• Update the Stata Guide
  • Continue the LaTeX rewrite
  • Take the decade-old (let's not wait for decades…) comments from Filip, Joël… into account
  • Finish writing the regression bits using Bittmann 2019 and Mehmetoglu and Jakobsen
• Update the course slides (let's stick with LaTeX despite love/hate)

Additional do-files

My many TODO files from 2017, (especially) 2018, 2019, 2020 have suggestions of extra do-files to create — shorter ones, ones that cover extra stuff beyond the scope of the course (e.g. merging, panel data).

I also have some very short "demo" do-files that I use in the first hour, as recaps of the previous session + introduction to the second hour of the current one.

Use that as an opportunity to…

• … include and demo more datasets?
• … document estout properly? (both for "Table 1" and regression tables)
• … rename the do-files, week01, week02… week12 for obsessive neatness?
  • have week0*-recap do-files with just the essentials too much work: streamline the week** ones
  • have xtra01 to xtra12 -- one 'bonus' do-file per week (see below)

Bonus do-files (which will move out some stuff from the main ones, and will cover some intermediate/advanced topics):

1. xtra01-pca -- plot a map + demo PCA (see below)
2. xtra02-merge -- download additional data from online + merge
3. xtra03-svy -- survey weights: WVS 99-04
4. xtra04-bootstrap -- survey weights: NHIS 2017 (repeat?) + bootstrap
5. xtra05-export -- export descriptive stats with estout
6. xtra06-tests -- survey weights: ESS 2008 (repeat?) + other association tests with ranks
7. xtra07-ts -- QOG time series with (extract of) 2023 edition? (serial correlation)
8. xtra08-panels -- robust and clustered SEs, fixed and random effects with QOG time series
9. xtra09-export -- export regression results with estout
10. xtra10-logit -- AUC/ROC, predicted probabilities, ordinal logit, multinomial (?)
11. xtra11-mfx -- marginal effects, bootstrap (already there at end, remove)
12. xtra12-count -- survey weights: GSS + neg binomial, count, Poisson etc.?

PCA example:

pca popgrowth-safewater
scoreplot, ms(i) mlab(country)
// note: tried using `kountry` to convert country names, failed so far
loadingplot
// demo arch effect, no strong 2nd dimension
pca lexp-safewater
scoreplot

Leaves out:

• MCA
• quantile regression, L1 (lasso), L2 (ridge)
• bootstrapped SEs in models
• Bayesian models
• multilevel models

Beyond teaching

I once considered publishing the Stata Guide, but publishing a Stata Guide, even though some publishers would take it, sounds bizarre in 2021. R is the current standard, with Julia and Python probably coming next or along.

• At least look at LeanPub, like Roger D. Peng
• Ask PSIA or the Presses de Sciences Po about it
• Go for Sage, like some kind of updated, no-menus Mehmetoglu and Jakobsen?