zmjones

there are just other tools that will get abused. bayes factors for example

how can an R package be "black boxy"? just look at the source. all the rf packages I am aware of (randomForest, party, and randomForestSRC) are well documented.

one of the coolest political science articles i've read. they have written several other papers on the topic. there most famous article on the topic is:

http://gking.harvard.edu/files/gking/files/censored.pdf

Meta-analysis is usually suspect.

Mearsheimer is an embarrassment to social science.

This is exactly why I prefer R for static maps. Would have taken like a quarter of the time, if that.

Here is mine: http://www.zmjones.com/mac-setup.html

this needs mathjax

I don't work at the office most of the time. I know thinkpads have good battery life generally, but I know there have been some battery life issues with linux in the past.

How is the battery life?

The nav bar and the commits page are too big I think. There ends up being a lot of space wasted. The commits page especially so. 5 commits should easily fit on a page. There are also two date information fields. One listing how many days ago the commit occurred and the other indicating what date the commit occurred on.

I would really like access to this too. I have been using sharelatex and the git support is still not there. My email is zmj@zmjones.com

I am hopeful that this works, as my collaborators can definitely not be bothered to learn git. The UI elements are way too big IMO.

You can't do much public good if you don't get re-elected. Since public opinion suggests people want more services and less taxes, something is going to have to give.

The funniest part was where he said politicians were not self-interested.

do want.

This is hardly unique to NYU. This is not uncommon in undergraduate courses on terrorism.

Definitely agree. Do check out Stan though. HMC is pretty fast compared to BUGS/JAGS.

You know github gives free private repositories to students right?

Why pie charts dude?

No the "strongest" don't necessarily survive. Evolution is only about individual survival insofar is a longer lifespan may result in more reproduction. The best adapted species in an particular environmental niche will outlast competitors.

And this is why we don't do statistical programming in Ruby.

Then maybe they should, I don't know, update their apps?

Yea I ended up re-writing my commit history to fix this.

I found a link to the 10.8 tools somewhere on SO. It was that or build everything from source...

their way of dealing with command line tools is rather annoying in this release

this seems way better than anything available on OSX at the moment.

Almost all of the empirical work done by political scientists and economists is on aggregate human behavior. Yes, the predictive power of the social sciences is often very weak, but social systems are incredibly complex, and in many ways vastly more difficult to understand than physical systems.

Looks really nice!