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

st: RE: graphing a series of vectors (latent growth curve model estimate summaries)

From   "Nick Cox" <>
To   <>
Subject   st: RE: graphing a series of vectors (latent growth curve model estimate summaries)
Date   Wed, 29 Oct 2008 13:12:59 -0000

. help twoway pcarrow 

Dopey example: 

sysuse auto
gen mpg2 = mpg + 2 
gen weight2 = weight - 100 
twoway pcarrow mpg weight mpg2 weight2

In other words, you must construct x and y variables specifying the
start of each arrow and similarly specifying the end of each arrow. 

See also 

SJ-5-2  gr0015  . . . . . . . . Stata tip 21: The arrows of outrageous
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N.
J. Cox
        Q2/05   SJ 5(2):282--284                                 (no
        tip for using graph twoway pcarrow for graphing changes
        over time

That Tip is accessible to all under the three-year window. 


Stephen P. Jenkins

A colleague wishes to summarise her estimates from a latent growth
curve model. I seek tips about how to construct graphs in the format
often used in this context.

In the graphs, the horizontal axis shows age. The vertical axis shows
the predicted n-year trajectory for a latent outcome. Plotted for each
of several starting ages is a vector or arrow -- a straight line
summarising the estimated trajectory going from the start year t to
the end year t+n with an arrowhead at the latter end. So, the graph is
the collection of vectors, and the vectors looked at collectively
themselves indicate a lifecourse trajectory (since number of years
used in each vector is relatively small, and 5-10 vectors may be used
to cover the whole life course).

An example is shown as "Figure 1: Predicted 8-year aging vectors of
physical impairment by educational attainment and family income", in
Jinyoung Kima and Emily Durden, "Socioeconomic status and age
trajectories of health", Social Science & Medicine 65 (2007) 2489-2502

The substantive medical application is not the essence of my question.
My questions concerns how to draw graphs like this, and related issues
such as data set-up. -graph twoway pccapsym- discussed on [G] p. 259,
and -graph twoway pcspike-, [G] p.268, seem to offer related
facilities but not quite what is requested here.

*   For searches and help try:

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