Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: RE: graph for propensity scores

From   "Nick Cox" <>
To   <>
Subject   st: RE: RE: graph for propensity scores
Date   Tue, 12 Apr 2005 14:57:22 +0100

Upwards and downwards designs appear popular 
for no good reason. In effect the reader is 
expected to be able to pick up one length, 
transfer it and superimpose it, all in one's
head, upon another length. Why this should be easier 
or more effective than comparing juxtaposed lengths 
beats me. 

The same issue arises with left and right 
(side-by-side) designs such as population 

In each case, small and subtle differences 
could easily be of interest or importance. 

The problem with histograms is naturally the 
loss of detail produced by binning. Often 
this is unimportant but frequently a researcher
does want to be sure that is so. 

To compare two sets of values qua distributions, 
-qqplot- is pretty nearly an optimal plot. To 
compare them as paired values, there are several 
good methods. A fairly lengthy discussion with 
references is given in 

Graphing agreement and disagreement. 
Stata Journal 4(3): 329--349 (2004)


David Harrison
> You could do that something like, e.g.
> sysuse auto
> twoway__histogram_gen length if !foreign, start(140) w(10) 
> freq gen(h0 x0)
> twoway__histogram_gen length if foreign, start(140) w(10) 
> freq gen(h1 x1)
> replace h1=-h1
> gen zero=0
> twoway (rbar h0 zero x0, barw(10)) (rbar h1 zero x1, barw(10)), ///
>     yla(-10 "10" -5 "5" 0(5)15) yti(Frequency) ///
>     legend(label(1 "Domestic") label(2 "Foreign"))
> Note the very handy command -twoway__histogram_gen- (two 
> underscores in the first gap!) that is not in the Stata 
> manuals but does have a help file.

Hans J. Baumgartner

> I am estimating treatment effects using propensity scores.
> To discuss the common support I’d like to graph the scores for the 
> treated and the controls.
> However, my graph knowledge is very limited and is only sufficient to 
> generate a graph, where the scores for the treated and the 
> controls are 
> displayed next to each other. That is I am using the by(.) option and 
> the command historgram.
> I’ve seen the scores are displayed upwards for the treated 
> and mirrored 
> downwards for the controls in one single graph.

*   For searches and help try:

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