Category Archives: R statistics

Stata tip: Doing something conditional on existence of a variable in the dataset, using a local

In a program I am writing I want the same utility file to be run on different datasets and certain variables in the dataset. But the datasets have different sets of variables. So a nice Stata tip, adapted for my … Continue reading

Posted in R statistics | Leave a comment

What students learning R in Econ 41-42 can aspire to…

The ideal job for every Econ major! You can follow Levin on Twitter here. Matt Levin is the data dude for CALmatters. His work entails distilling complex policy topics into easily digestible charts and graphs, finding and writing original stories … Continue reading

Posted in R statistics

Is there a good way to download the coordinates of a path or route on google map into a file?

The folks at stackoverflow just dismissed this question. But is an honest and useful question to have answered, for some applications.  And here is a good ttip from fentonsrule: This is something I learned on my own. If you create … Continue reading

Posted in R statistics

The perfect graph for every undergraduate statistics class, from Financial Times

I object to usage of “trend” to indicate “relationship” (for me “trend” implies over time) but that is just a quibble.  Perfect illustration of importance of visualizing data. PS. Doug Campbell reminded me of importance of a title for the … Continue reading

Posted in Development thinking, R statistics

Reading in PDF files into R to do text analysis

Turns out Clay Ford at the University of Virginia wrote a nice tutorial for this, and a package does the trick very nicely.  I tested the “update” at the bottom of the post which shows how to use the pdftools … Continue reading

Posted in R statistics

Stata gets a markdown package… will try this in coming weeks

Rigorous documentation of the analysis plan, procedure, and computer codes enhances the comprehensibility and transparency of data analysis. Documentation is particularly critical when the codes and data are meant to be publicly shared and examined by the scientific community to … Continue reading

Posted in R statistics

Non-citizens voting? Wonderful straightforward analysis from Ansolabehere, Luks, and Schaffner

Stepping back from the immediate question of whether the CCES in fact shows a low rate of voting among non-citizens, our analysis carries a much broader lesson and caution about the analysis of big databases to study low frequency characteristics … Continue reading

Posted in R statistics