Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: RE: graphing question - how to smooth lines connecting observations


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: graphing question - how to smooth lines connecting observations
Date   Wed, 14 Apr 2010 12:23:04 +0100

The reference is presumably to some feature in MS Excel.

Martin has a serious point. Many on this list don't have access to other
programs which list members refer to, or don't have detailed knowledge
of such other programs, regardless of what individuals regard as
standard or routine in their own institution or discipline. For example,
many people, especially among the more active answerers of questions on
this list, have little or no acquaintance with MS Office. 

I think I have seen graphs produced this way in student work. I don't
know, and don't want to know, precisely how Excel does it. The problem
is that _connecting_ the lines with a _smooth_ curve is almost always a
bad idea. Unless the number of points is very small the curve will
display spurious fluctuations as a result of overfitting. 

For what you want, -twoway mspline- is probably your simplest option in
Stata. More serious smoothing can be done in many different ways. 

Nick 
n.j.cox@durham.ac.uk 

Martin Weiss

Not all Statalisters may want to speculate on what other applications
do,
and your misspelling their name may further discourage answers. So the
way
forward for you is to spell out clearly, preferably using a dataset
shipped
with Stata, what you want your graph to look like/which features you
dislike
and want to change. Take as an example this one:

*******
sysuse auto, clear
gen byte number=_n
tw (connected displacement number, sort)
*******

Zamira S

I am trying to generate a graph with smooth-looking lines connecting the
observations. In excell it is an option but I am trying to avoid going
back
and force between excell and stata. Any suggestions would be greatly
appreciated. Here is what I have:
 
twoway (connected growth timeline, lcolor(black) lpattern(solid)
lwidth(medthin) msymbol(circle) mcolor(black)), ytitle(GDP per capita
growth) yline(0 2, lwidth(vvvthin) lcolor("192 192 192")) xtitle(Time)
xlabel(-3(1)9) xline(-0.1 1.1 5.1, lcolor("87 87 87") lpattern(dash))
title(GDP Per Capita Growth (%)) subtitle(Constant US$) saving(growth,
replace)
 

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index