Twoways come to mind. The first is to -reshape- the data into a long format
and then use -xtline-. The second is to -xpose- the data and use -twoway
connect- Examples below:
Example 1.
. l
+---------------------------+
| patient value1 value2 |
|---------------------------|
1. | 1 51 48 |
2. | 2 49 53 |
3. | 3 47 59 |
+---------------------------+
. reshape long value, i(patient) j(time)
(note: j = 1 2)
Data wide -> long
----------------------------------------------------------------------------
-
Number of obs. 3 -> 6
Number of variables 3 -> 3
j variable (2 values) -> time
xij variables:
value1 value2 -> value
----------------------------------------------------------------------------
-
. tsset patient time
panel variable: patient, 1 to 3
time variable: time, 1 to 2
. xtline value, overlay plopts1(recast(connected))
plopts2(recast(connected)) plopts3(recast(connected)) xlabel( 1 2)
Or, an alternative way to get the dots at each point
. xtline value, overlay xlabel( 1 2) plot(scatter value time) legend(order(1
2 3))
Example 2
. l
+--------------------+
| var1 var2 var3 |
|--------------------|
1. | 1 51 48 |
2. | 2 49 53 |
3. | 3 47 59 |
+--------------------+
*Make sure the data are saved!
. xpose, clear
. drop in 1
(1 observation deleted)
. gen time = _n
. forv i = 1/3 {
2. label var v`i' "Patient `i'"
3. }
. twoway (connect v1 time) (connect v2 time) ( connect v3 time) , xlabel(1
2)
Hope this helps,
Scott
----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Friday, April 16, 2004 5:19 PM
Subject: st: simple (?) graph question
> All,
>
> I have a simple graph question. I paged through the Graphics manual and
> looked at the pull down menu but can't figure out an answer.
>
> I have a small data set of before and after lab values on 20 individuals.
> I would like to do a dot graph with two values on the x-axis, "before" and
> "after", and the lab value on the y axis. I would like to connect the
> before and after dots for each individual with a line, so I would have 20
> lines.
>
> The data are arranged:
>
> Patient # Value 1 Value 2
> 1 51 48
> 2 49 53
> 3 47 59
> .
> .
> .
> and so on...
>
> Thanks, and have a good weekend,
> Jim Seward
*
* 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/
