Convert SAS code into SPSS syntax

[Lily]

Hey guys, I am new to SPSS macro. Now I have some SAS code and try to
convert them to SPSS syntax.It is really an emergency. Could anybody
help me out? The SAS code are as follows:

1. %let a=10;

data new;
set old;
where b=%eval(&a-1);
run;

2.
%macro byweek(a, b);
proc summary data=&a.week;
by system cat name;
var absent enrolled;
output out=&b._out sum=&b._absent &b._enrolled;
run;

data &b._out2;
set &b._outs;
&b._rate=(&b._absent*100)/&b._enrolled;
keep system cat name
&b._absent &b._enrolled &b._rate;
run;

data &b._out3;
merge &b._out2 list;
by system cat name;
run;

%mend;

Thank you so much .

[Bruce Weaver]

The folks most likely to help probably don't know SAS very well, if at
all. So it would help if you added comments explaining what the SAS
code is doing.

--
Bruce Weaver
bwe...@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

[Lily]

Thanks for your reply Bruce. Well, the description for the SAS code is
as follos:

1. %let a=10: creates a macro variable "a" and assign value 10 to
this marco variable.
b=%eval(&a-1): %eval makes the arithmetic possible. it equals b=
(10-1)=9

2. Create a new data set new based on the old data set.
&b: refers to macro variable b.
Proc summary: we can generate the summary statistics using it

I tried to convert this SAS code to the SPSS syntax and got the
following SPSS syntax. But I don't know how to refer the macro
varibale in SPSS, so I am wondering if you can replace the &b with
appropriate spss syntax. Please correct me if I made some mistakes.

DEFINE !byweek(a, b)

GET FILE=”F:\ &a.week”.
SORT CASES BY system cat name.

AGGREGATE
/OUTFILE= * MODE=ADDVARIABLES
/BREAK=system cat name
/&b._labsent &b._enrolled =SUM(absent enrolled)
COMPUTE &b._rate=(&b._absent*100)/&b._enrolled.

SAVE OUTFILE='C: &b._out2school.sav'
/KEEP= system cat name &b._absent &b._enrolled &b._rate.

GET FILE='F:\listschool.sav'.
SORT CASE BY system cat name.
GET FILE='F:\&b._out2school.sav'.
SORT CASE BY system cat name.
MATCH FILES/FILE=*
/FILE='name'
/BY system cat name.
EXECUTE.

!ENDDEFINE.
!byweek a=current b=curr.
!byweek a=previous b=prev.

3.How can we connect SPSS results with MS-Excel using DDE (Dynamic
data exchange).

Thanks again

[Bruce Weaver]

On May 11, 11:35 pm, Lily <pointbreak3...@gmail.com> wrote:
> Thanks for your reply Bruce. Well, the description for the SAS code is
> as follos:
>
> 1. %let a=10:  creates a macro variable "a" and assign value 10 to
> this marco variable.
>    b=%eval(&a-1):   %eval makes the arithmetic possible. it equals b=
> (10-1)=9

There is no SPSS equivalent of %eval AFAIK. One has to resort to
tricks to do arithmetic in the macro language. I believe there are
some examples on Raynald's website (spsstools.net), and probably in
the archives of this group.

>
> 2.   Create a new data set new based on the old data set.
>      &b: refers to macro variable b.
>      Proc summary:  we can generate the summary statistics using it
>
> I tried to convert this SAS code to the SPSS syntax and got the
> following SPSS syntax. But I don't know how to refer the macro
> varibale in SPSS, so I am wondering if you can replace the &b with
> appropriate spss syntax. Please correct me if I made some mistakes.

I don't have a lot of time, but here are a couple observations (see
below).

>
> DEFINE !byweek(a, b)

DEFINE !byweek (a=!TOKENS(1) /
b=!TOKENS(1))

>
> GET FILE=”F:\ &a.week”.
> SORT CASES BY system cat name.
>
> AGGREGATE
> /OUTFILE= * MODE=ADDVARIABLES
> /BREAK=system cat name

> /&b._labsent &b._enrolled =SUM(absent enrolled) .

COMPUTE is a separate command. You need the command terminator I added
to the previous line.

> COMPUTE &b._rate=(&b._absent*100)/&b._enrolled.

>
> SAVE OUTFILE='C: &b._out2school.sav'
> /KEEP= system cat name     &b._absent &b._enrolled &b._rate.
>
> GET FILE='F:\listschool.sav'.

DATASET NAME name.

* I assmume listschool.sav is the file you are trying merge below.

> SORT CASE BY system cat name.
> GET FILE='F:\&b._out2school.sav'.
> SORT CASE BY system cat name.
> MATCH FILES/FILE=*
>    /FILE='name'
>    /BY system cat name.
> EXECUTE.
>
> !ENDDEFINE.
> !byweek a=current b=curr.
> !byweek a=previous b=prev.
>
> 3.How can we connect SPSS results with MS-Excel using DDE (Dynamic
> data exchange).

I don't know.

>
> Thanks again

HTH.

[Lily]

Thanks Bruce, it does help a lot. Have a nice evening.

> bwea...@lakeheadu.cahttp://sites.google.com/a/lakeheadu.ca/bweaver/

[bbogda...@gmail.com]

what would be the spss equivalent of SAS retain statement ?

[Bruce Weaver]

On Friday, April 22, 2016 at 12:21:21 AM UTC-4, BOGDAN BOGDANOVIC wrote:
> what would be the spss equivalent of SAS retain statement ?

It is said that God and helps those who help themselves. The same might be said of members of this forum.

TRANSLATION: You will be far more likely to get help if you explain what "retain" does in SAS. Provide a small small example showing the current state of your data, and the desired state after "retain", for example. Or provide a link to a page that explains it. Don't make us do the work for you.

HTH.

[David Marso]

Bruce,
I answered this question 2 threads below (Either LEAVE or LAG depending upon context -which is not provided-.
BOGDAN ! please don't double post and follow up on your queries when asked for clarification.