st: new package -binscatter- available in SSC

[Michael Stepner <stepner@mit.edu>]

-binscatter- is a Stata routine which draws binned scatterplots, and
is now available in the SSC.

Binned scatterplots provide a non-parametric method for visualizing
the relationship between two variables. With a large number of
observations, a scatterplot that plots every data point would become
too crowded to interpret visually. binscatter groups the x-axis
variable into equal-sized bins, computes the mean of the x-axis and
y-axis variables within each bin, then creates a scatterplot of these
data points.  The result is a non-parametric visualization of the
conditional expectation function.

binscatter has built-in options to control for covariates before
plotting the relationship. It will also plot fit lines based on the
underlying data, and can automatically handle regression
discontinuities.  It is optimized specifically for speed in large
datasets.

For examples of existing work utilizing binscatter, see:

http://obs.rc.fas.harvard.edu/chetty/ret_savings.html
(first two figures on website, more in the paper)

http://obs.rc.fas.harvard.edu/chetty/value_added.html
(middle figure on website, more in the paper)

If you're interested, you can install the program via -ssc install binscatter-.

Best regards,
Michael
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/