Hi All,
I am working on a plot like this: https://blogs.sas.com/content/graphicallyspeaking/2018/03/02/series-plot-varying-attributes/
I am currently struggling with the datastep where I need to define that the last point of the previous curve segment is replicated as the first point of the next segment.
I have two dates defining the change in color. When age (age_measure) passes age 1, then I want a change in the color and again when age passes age 2 a new change in color.
So I need to get the determinant of the colorresponse variable appear twice: Meaning age 1 appears both as colorresponse variabel 1 and 2, and age 2 appearing both as colorresponse variable 2 and 3 to get the datapoints to connect in the plot.
Do you have any ideas on how to define that in a datastep?
Example data below:
patientid = patient id, a1 = age 1, a2 = age 2, visit date = examination date, measure = measurement at examination date, age_measure = age at measurement.
The curve is measure as a function of age_measure, where the curve needs to change color when age 1 and age 2 are passed (interpolated values of age_measure.). Age 1 always comes before age 2, but some patients jump directly to age 2. Some patients might not have either age 1 or 2.
I don't know if this this possible in proc sgplot as the curve needs to "break" several times, like in a spaghetti-plot. But maybe proc gplot works, I have just never used that before.
data patients;
input patientid $ a1 a2;
datalines;
1 32 34
2 23 .
3 22 35
4 30 31
5 . 28
;
data visits;
input patientid $ visit_date :yymmdd8. measure age_measure;
format visit_date yymmdd10.;
datalines;
1 20180101 3.4 20
1 20180505 2.3 35
2 20170210 7.3 15
2 20170217 7.2 24
2 20170220 7.1 35
3 20170221 5.4 33
4 20180202 3.4 30
4 20180204 3.2 37
5 20180504 5.6 28
5 20180505 5.0 32
;
data have;
merge
patients
visits
;
by patientid;
run;
