creating BOXPLOT for multiple variables

[radhikaa4]

Hi I would like to create a boxplot for multiple variables but not just 1 variable

ID	time_hour_1	time_hour_2	time_hour_3	time_hour_4
1	40	122	21	12
2	20	112	12	66
3	30	143	14	43
4	35	145	17	43
5	22	124	10	12

I tried Proc Boxplot data = a.test;

hbar time_hour_1 time_hour_2;

RUN;

Not working. thanks!

[Reeza]

Of course it doesn't, HBAR isn't listed under PROC BOXPLOT, HBAR stands for horizontal bar so not sure why you think that makes sense here either. 

Take a look at the documentation first, they have an example that's really close to yours. 

Note that the issue is the structure of your data. Determine how to change your data structure to match theirs and you can use the exact code in the documentation (hint: PROC TRANSPOSE). 

https://documentation.sas.com/?docsetId=statug&docsetTarget=statug_boxplot_examples02.htm&docsetVers...

Note that at the top of the page, there's a link to the full code if you want, which gets you the full program to run and play with. 

[Reeza]

Two different ways to do this, using SGPLOT instead.

data have;
    input ID time_hour_1 time_hour_2 time_hour_3 time_hour_4;
    cards;
1   40  122 21  12
2   20  112 12  66
3   30  143 14  43
4   35  145 17  43
5   22  124 10  12
;
    ;
    ;;
run;

data long;
    set have;
    array tm(*) time_hour_1 - time_hour_4;

    do i=1 to dim(tm);
        hour=compress(vname(tm(i)), 'kd');
        Value=tm(i);
        output;
    end;
    keep id hour value;
run;

proc sgplot data=long;
    vbox value / group=hour;
run;

proc sgplot data=have;
    vbox time_hour_1;
    vbox time_hour_2;
run;

---

@radhikaa4 wrote:

Hi I would like to create a boxplot for multiple variables but not just 1 variable

ID	time_hour_1	time_hour_2	time_hour_3	time_hour_4
1	40	122	21	12
2	20	112	12	66
3	30	143	14	43
4	35	145	17	43
5	22	124	10	12

I tried Proc Boxplot data = a.test;

hbar time_hour_1 time_hour_2;

RUN;

Not working. thanks!

---