Notes on slides for future

[gustavekenedi]

Lecture causality:

• slide 35: we actually need to state that we "assume" mean independence of e_i with respect to D_i. (in the simple linear regression lecture we only said that e_i and x_i were not correlated). only in that case can we get that E(e_i|D_i = 1) = 0.

[gustavekenedi]

Note to self: use datacamp for instructors so students can practice more on the weeks without quizzes (and learn other material on their own time if they're interested)? it's free for classes for 6 months at a time and seems pretty easy to set up. i've used datacamp a lot and have greatly benefited from it.

https://www.datacamp.com/groups/education

[gustavekenedi]

Very good exercises! https://github.com/skranz/RTutor

[gustavekenedi]

regression inference slides are a bit too long: less theory?

[gustavekenedi]

various (overwhelmingly non-economic) datasets: https://github.com/rfordatascience/tidytuesday

[floswald]

I think we should not use datacamp. https://www.noamross.net/2019/04/12/datacamp-sexual-assault/

[gustavekenedi]

Compilation of resources to learn R: https://github.com/iamericfletcher/awesome-r-learning-resources

[gustavekenedi]

https://datasciencebox.org/

[gustavekenedi]

https://pkg.garrickadenbuie.com/xaringanExtra/panelset/#1 pretty neat! not sure when would come in handy but putting it here in case

[gustavekenedi]

for Sciences Po logo on all slides, seems like a neat solution to our having to add code to have the logo: https://pkg.garrickadenbuie.com/xaringanExtra/#/logo

[gustavekenedi]

just discovered that xaringan has its in-house decktape function: xaringan::decktape(): https://www.garrickadenbuie.com/blog/print-xaringan-chromote/ will check it out later today, just writing this as a note to self. seems slightly more convenient than switching to the terminal for pdf conversion

[gustavekenedi]

For students after first lecture: https://gitlab.com/stragu/DSH/-/blob/master/R/terminology.md