Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: error bars


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: error bars
Date   Thu, 19 Jun 2008 18:07:56 +0100

Several issues have been raised in this thread, principally what is best
to do and how to do it. 

I dislike "detonator plots" (Paul Seed's name?), a.k.a. "dynamite plots"
(Frank Harrell's name?), as heartily as anyone else, but let that pass. 

Moleps here asserts that -ciplot- (one of my programs on SSC) produces
two identical error bars for two different variables. 

I can't see Moleps' data from here but I believe that assertion to be
false, unless of course the variables happen to collapse to the same
confidence intervals. 

sysuse auto 
ciplot turn mpg, by(foreign)

is one simple example to the contrary. 

In another posting, Moleps makes the same assertion about -cibplot-
(also mine on SSC), and the same rebuttal applies. 

-cibplot- was written a while back in 2003 in response to a similar
thread. It was a quick hack. -cibplot- incorporates a prejudice of mine
in that error bars are drawn on top of estimate bars and so cannot be
overwritten, except by devious means. 

However, it has taken me five years to realise that -cibplot- was
unnecessary all along.

The -recast()- option applied to -ciplot- produces estimate bars rather
than estimate point symbols. 

ciplot turn mpg , by(foreign) recast(bar) 

The usual rhetoric here is start bars at 0 so that error bars look
trivially small: 

ciplot turn mpg , by(foreign) recast(bar) yla(0(10)40)

As old hands will realise, this is not a matter of -ciplot- being
especially smart. Rather, if I see far, I am standing on StataCorp's,
specifically Vince Wiggins' shoulders. (He'd rather I dismounted
promptly.) 

But -ciplot- will get no further development from me. My affections have
transferred to -stripplot- (also on SSC), which shows the raw data too. 

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

moleps islon

I' making a double error bar graph- e.g two different variables x and
y with error bars categorised according to z. I've tried using ciplot
x y, by(z) however I get a graph with two identical error bars (though
different symbol, but identical values) for each instance of z. Any
idea how to remedy this? If I need to make bars (a bar graph)  for
each category and then have the positive part of the error bar
projecting from the top- how do I go about this?

On Mon, Jun 16, 2008 at 2:23 PM, Nick Cox <n.j.cox@durham.ac.uk> wrote:

> Standard errors can come from lots of places, including the -ci-
> command.
>
> Moleps seems to be implying that -ci- does the right calculation for
his
> or her purposes.
>
> (I note that Moleps has yearly results, but -ci- for separate years
does
> nothing about any time series structure in the data, for example about
> taking serial correlation into account.)
>
> -ciplot- and -stripplot- from SSC both do -ci-type calculations and
> graphing in one.

 moleps islon <moleps2@gmail.com> wrote:
>> I need to make a an error-bar graph categorised by year. I've tried
>> using serrbar mean(x) etc, but I cant find a command for the standard
>> error. Do I need to run a CI, generate a new variable from the ci
>> result and feed that into serrbar?
>

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index