Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: new bivariate density estimation and graphics routine

From   Steve Samuels <>
Subject   Re: st: new bivariate density estimation and graphics routine
Date   Thu, 27 Dec 2012 14:39:55 -0500

Oops, naturally you meant the "[]" to indicate that the scatter options were, indeed,

Thanks, John & Kit!

In the -help-, the  quoted option "scatter[(scatter_options)]" appears to be incorrect.

e.g. in the auto data set:

. bidensity weight trunk, scatter[(mlabel(rep78))]
option [() not allowed

but omitting the square brackets works.

.bidensity weight trunk, scatter(mlabel(rep78))


On Dec 27, 2012, at 2:10 PM, Christopher Baum wrote:

. ssc desc bidensity

package bidensity from

    'BIDENSITY': module to produce and graph bivariate density estimates


     bidensity produces bivariate kernel density estimates and graphs
    the result using  a twoway contourline plot, optionally
    overlaying a scatterplot.  The default kernel is Epanechnikov;
    all of the kernels provided by -kdensity- are also available.
    Compared to Baum's -kdens2- (SSC), which was recently enhanced to
    produce contourline plots,   -bidensity- computes the bivariate
    kernel densities much more efficiently through use of Mata, and
    provides a choice of kernel estimators. The estimated densities
    can be saved in a Stata   dataset or accessed as Mata matrices.

    KW: density estimation
    KW: bivariate density
    KW: contourline plots

    Requires: Stata version 12.1

    Distribution-Date: 20121227

    Author: John Luke Gallup, Portland State University
    Support: email

    Author: Christopher F Baum, Boston College
    Support: email

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index