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: using xtline to create panel data plots
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: using xtline to create panel data plots
Date
Tue, 13 Sep 2011 19:28:28 +0100
Even if there were, a chaos of different scales would surely ensue.
What I suggest is that you think of
1. whether some nonlinear scale would help (seemingly not logarithmic
if 0 is a value, but cube root or asinh are other possibilities).
2. an appropriate standardisation for graphical purposes, which might
be for each panel (e.g.)
(value - mean) / sd, value / mean or (value - min)/(max - min)
With #1 or #2, you'd need to create the variable first: -gen- would
be handy for #1 and -egen- would be handy for #2.
Nick
On Tue, Sep 13, 2011 at 7:18 PM, Venkiteshwaran, Vinod
<[email protected]> wrote:
> I have a panel data set for several industries spanning a twenty year time period.
> The variable I am trying to plot is the value of a financial ratio across time for each industry.
> The range of this ratio value differs by industry.
> For eg. industry-1 has a range of (0 .0068227, 0.0537253) while industry-2 has a range of (0.0102397 , 0.2309189) for the same ratio.
> The overall sample range for this variable is (0, 0 .237276)
> Is there anyway to get xtline to plot graphs for each industry panel using the range applicable to only that industry as opposed to the range for the entire sample.
> I understand I can create the graphs one industry at a time to resolve this issue.
*
* 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/