sum function in SAS

[frakje]

Hello

One simple question.

I want to calculate the sum of a variable in my long formatet dataset.  

My dataset looks like this:

ID        gmfcs_dummy  

1000    0

1000    1

1000    1

2000    0

2000    0

2000    0

3000    1

3000    1

3000    1

I want to calculate the sum of gmfcs_dummy so it sums up in rising order like this:

ID         gmfcs_dummy   gmfcs_dummy2

1000            1                       1

1000            1                       2

2000            0                       0

2000            0                       0

2000            0                       0

3000           1                        1

3000           1                        2

3000           1                        3

How can I do this?

Kind regards Frank

[ed_sas_member · Accepted solution]

Hi @frakje 

You can use this:

data have;
	input ID gmfcs_dummy;
	datalines;
1000 0
1000 1
1000 1
1000 0
2000 0
2000 0
2000 0
3000 1
3000 1
3000 1
;
run;

data want;
	set have;
	by ID;
	if first.ID then gmfcs_dummy2 = 0;
	if gmfcs_dummy=1 then gmfcs_dummy2 + gmfcs_dummy;
	else gmfcs_dummy2 = 0;
run;

[Kurt_Bremser]

---

@frakje wrote:

Hello

One simple question.

I want to calculate the sum of a variable in my long formatet dataset.  

My dataset looks like this:

ID        gmfcs_dummy  

1000    0

1000    1

1000    1

2000    0

2000    0

2000    0

3000    1

3000    1

3000    1

I want to calculate the sum of gmfcs_dummy so it count upwise like this:

ID         gmfcs_dummy   gmfcs_dummy2

1000            1                       1

1000            1                       2

2000            0                       0

2000            0                       0

2000            0                       0

3000           1                        1

3000           1                        2

3000           1                        3

How can I do this?

Kind regards Frank

---

Why does your intended result contain only 2 observations for ID 1000, but 3 for ID 2000?

[frakje]

Hallo Kurt

It's because the observations/patients has an new treatment every year. Each patient can have different entry dates in the follow-up program.

Patient with ID 1000 have e.g. first treatment in 2016, second treatment in 2017 and third treatment in 2018. Therefore there is three records for that specific patient.

[Kurt_Bremser]

But 1000 has only 2 (TWO) observations in your result.

[frakje]

Oh, sorry!

Typing error. It should have been three

[Kurt_Bremser]

Then try this:

data want;
set have;
by id;
if first.id
then gmfcs_dummy2 = gmfcs_dummy;
else gmfcs_dummy2 + gmfcs_dummy;
run;