Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Multiple density plots, rotated and distributed on x-axis
From
Venable <[email protected]>
To
[email protected]
Subject
Re: st: Multiple density plots, rotated and distributed on x-axis
Date
Thu, 4 Aug 2011 16:11:45 -0400
Scott and Austin,
Both of those are extremely helpful, thank you both very much.
Scott's is a bit closer to what I had in mind but vioplot is quite
impressive as well.
On Tue, Aug 2, 2011 at 3:06 PM, Austin Nichols <[email protected]> wrote:
> Venable <[email protected]>:
> I love Scott's example, but see also -vioplot- on SSC:
> sysuse auto, clear
> vioplot mpg, over(foreign)
>
> On Tue, Aug 2, 2011 at 1:55 PM, Scott Merryman <[email protected]> wrote:
>> Like this?
>>
>> clear
>> sysuse auto
>> set obs 5000
>> kdensity mpg if fore == 0, n(5000) k(gauss) bw(3) gen(h0 d0)
>> line d0 h0
>> kdensity mpg if fore == 1, n(5000) k(gauss) bw(3) gen(h1 d1)
>>
>> replace d1 = d1 + .2
>>
>> tabstat mpg, by(fore) stat(p25 p50 p75) nototal
>>
>> line h1 d1 || line h0 d0, xtitle("") xlabel( .02 "Domestic" .22
>> "Foreign" ) ///
>> legend(off) scheme(lean1) ///
>> || scatteri 24.5 .264 19 .077 , recast(connected) ms(O) lw(thin) ///
>> || scatteri 21 .255 16.5 .0698 , recast(connected) ms(O) lw(thin) ///
>> || scatteri 28 .2427 22 .0601 , recast(connected) ms(O) lw(thin)
>>
>>
>> Scott
>>
>>
>> On Mon, Aug 1, 2011 at 1:26 PM, Venable <[email protected]> wrote:
>>> Dear Statalist,
>>>
>>> I would like to create a graph that contains several kernel density
>>> plots, but with the plot rotated 90% right from the usual orientation,
>>> and with each plot aligned to a point on the x axis.
>>>
>>> To be more concrete (borrowing an example from Angrist and Pischke,
>>> Mostly Harmless Econometrics, Figure 3.1.1), suppose I wanted to show
>>> the distribution of incomes in the US at different levels of eduction,
>>> e.g. 8 years, 12 years, 14, 16, etc. The x-axis of the graph would be
>>> years of education and the y axis the level of income. Above each
>>> year, there would be a kernel density plot of incomes for that year,
>>> with each rotated to the right, so that the "height" (density) is
>>> horizontal distance.
>>>
>>> An additional bonus would be to be able to mark a few quantiles on
>>> each density and connect these between years. For example, mark the
>>> 25th, median and 75th percentile for each year and have these points
>>> connected.
>>>
>>> Is this possible in Stata?
>>>
>>> I fear I may not be explaining this well in words, in spite of the
>>> fact that it is very easy to show as a picture. Unfortunately Figure
>>> 3.1.1 is not available in the Google Preview version of Mostly
>>> Harmless Econometrics, so I have posted a simple hand-drawn version
>>> here:
>>>
>>> http://dl.dropbox.com/u/2937392/density-plot-sketch.pdf
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/