I have a very short (<20 rows) data set that looks like:
Effective_Date Pct
-------------- ---
01JAN2000 50%
...
11FEB2014 55%
13JUL2014 65%
I'd like to write a macro which takes a date, Eval_Date and returns the Pct which was effective on that date. To be clear, I know that this can be done with some kind of PROC SQL construction, but I want to write a function-style macro that can be used in the data step.
For example, %pct('12jul2014'd) should evaluate to 55%.
Assuming your source dataset is effpct and pct is numeric, formatted as percent., use it to create a format containing every day and the effective percent:
/* Merge without a by statement, using firstobs=2 to do a look-ahead join to
determine the 'effective to' date */
data pct_fmt ;
retain fmtname 'EFFPCT' type 'N' ;
merge effpct
effpct (firstobs=2 keep=effective_date rename=(effective_date=to_date)) ;
if missing(to_date) then to_date = date() ; /* Take last record up to current date */
do start = effective_date to (to_date - 1) ;
label = pct ;
output ;
end ;
run ;
/* 'Compile' the format */
proc format cntlin=pct_fmt ; run ;
/* Abstract put(var,format) into a function-style macro */
%MACRO PCT(DT) ;
put(&DT,EFFPCT.) ;
%MEND ;
/* Then use it in a datastep... */
data want ;
input date date9. ;
eff_pct = %PCT(date) ;
format eff_pct percent9. ;
datalines ;
01JAN2000
13FEB2014
20JUL2014
;
run ;
Or alternatively, use %SYSFUNC and putn to be able to convert a date to percent outside of a datastep, e.g. in a title statement :
%MACRO PCT2(DT) ;
%SYSFUNC(putn(%SYSFUNC(putn(&DT,EFFPCT.)),percent.))
%MEND ;
title "The effective pct on 09JUL2013 was %PCT2('09jul2013'd)" ;
Related
I'm trying to import csv file to SAS using proc import; I know that guessingrows argument will determine automatically the type of variable for each column for my csv file. But there is an issue with one of my CSV file which has two entire columns with blank values; those columns in my csv file should be numeric, but after running the below code, those two columns are becoming character type, is there any solutions for how to change the type of those two columns into numeric during or after importing it to SAS ?
Here below is the code that I run:
proc import datafile="filepath\datasetA.csv"
out=dataA
dbms=csv
replace;
getnames=yes;
delimiter=",";
guessingrows=100;
run;
Thank you !
Modifying #Richard's code I would do:
filename csv 'c:\tmp\abc.csv';
data _null_;
file csv;
put 'a,b,c,d';
put '1,2,,';
put '2,3,,';
put '3,4,,';
run;
proc import datafile=csv dbms=csv replace out=have;
getnames=yes;
run;
Go to the LOG window and see SAS code produced by PROC IMPORT:
data WORK.HAVE ;
%let _EFIERR_ = 0; /* set the ERROR detection macro variable */
infile CSV delimiter = ',' MISSOVER DSD lrecl=32767 firstobs=2 ;
informat a best32. ;
informat b best32. ;
informat c $1. ;
informat d $1. ;
format a best12. ;
format b best12. ;
format c $1. ;
format d $1. ;
input
a
b
c $
d $
;
if _ERROR_ then call symputx('_EFIERR_',1); /* set ERROR detection macro variable */
run;
Run this code and see that two last columns imported as characters.
Check it:
ods select Variables;
proc contents data=have nodetails;run;
Possible to modify this code and load required columns as numeric. I would not drop and add columns in SQL because this columns could have data somewhere.
Modified import code:
data WORK.HAVE ;
%let _EFIERR_ = 0; /* set the ERROR detection macro variable */
infile CSV delimiter = ',' MISSOVER DSD lrecl=32767 firstobs=2 ;
informat a best32. ;
informat b best32. ;
informat c best32;
informat d best32;
format a best12. ;
format b best12. ;
format c best12;
format d best12;
input
a
b
c
d
;
if _ERROR_ then call symputx('_EFIERR_',1); /* set ERROR detection macro variable */
run;
Check table description:
ods select Variables;
proc contents data=have nodetails;run;
You can change the column type of a column that has all missing value by dropping it and adding it back as the other type.
Example (SQL):
filename csv 'c:\temp\abc.csv';
data _null_;
file csv;
put 'a,b,c,d';
put '1,2,,';
put '2,3,,';
put '3,4,,';
run;
proc import datafile=csv dbms=csv replace out=have;
getnames=yes;
run;
proc sql;
alter table have
drop c, d
add c num, d num
;
I'm having some trouble referencing a global macro variable outside of the macro to create a new data set. The global variable was created to run a loop for creating several yearly data sets using a vector of specified years, as you can see in the code below:
%macro loopyear;
%global year;
%do year = 2004 %to 2017;
proc import datafile = "C:\Filepath\blah.txt"
dbms = dlm out = blah&year.; /*Creates a dataset for each year, e.g. blah2004, blah2005, etc.) */
delimiter = " ";
getnames = no;
run;
data blah&year.;
set blah&year.;
year = &year.;
run;
proc sql;
create table blah&year._rail as
select year, var1, var2, var3, var4
from blah&year.
where var2= "rail";
quit;
%end;
%mend loopyear;
%loopyear;
/*Merge all year datasets into one master set*/
data blah_total;
set blah&year._rail;
run;
When I try to create the master data set outside of the macro, however, I get the following error:
data blah;
set blah&year._rail;
run;
ERROR: File work.blah2018_rail.data does not exist
This is frustrating because I'm only trying to create the master set based on 2004-2017 data, as referenced in the macro variable. Can someone help me pinpoint my error -- is it in the way I defined the global variable, or am I missing a step somewhere? Any help is appreciated.
Thanks!
This is an interesting quirk of both macro and data step do-loops in SAS - the loop counter is incremented before the exit condition is checked, so after your loop has run it will be one increment past your stop value, e.g.:
%macro example;
%do i = 1 %to 3;
%put i = &i;
%end;
%put i = &i;
%mend;
%example;
Output:
i = 1
i = 2
i = 3
i = 4
For your final step you probably want the set statement to look like this:
set blah2004_rail ... blah2017_rail;
You could write a macro loop to generate the list and move the data step inside your macro, e.g.
set %do year = 2004 %to 2017; blah&year._rail %end;;
The second semi-colon is important! You need one to close the %end and one to terminate the set statement.
Change your naming structure. Have a common prefix and put the year at the end, then you can use the semi colon to short reference all the datasets at once.
%macro loopyear;
%global year;
%do year = 2004 %to 2017;
proc import datafile = "C:\Filepath\blah.txt"
dbms = dlm out = blah&year.; /*Creates a dataset for each year, e.g. blah2004, blah2005, etc.) */
delimiter = " ";
getnames = no;
run;
data blah&year.;
set blah&year.;
year = &year.;
run;
proc sql;
create table blah_rail_&year. as
select year, var1, var2, var3, var4
from blah&year.
where var2= "rail";
quit;
%end;
%mend loopyear;
%loopyear;
/*Merge all year datasets into one master set*/
data blah_total;
set blah_rail: ;
run;
I wonder if there is a way to query in SAS to select a subgroup, just like select option in Postgres
SELECT *
FROM s.diagnoses
WHERE icd9code = ANY ('{2910,2911,2912,2913,2914,2915,3456,3457,3458}');
Also is there way to specify ranges instead of the actual value eg: between 2910-2915
The diagnosis codes are characters not numeric. I am using the SAS University Edition.
In case you want to specify range then you have to convert the character field into numeric and then give the range
/***** if you want to mention each icd9code*****/
data have;
set diagnoses (where=(icd9code in ('2910' '2911' '2912' '2913' '2914' '2915' '3456' '3457' '3458')));
run;
/***** if you want to give range *****/
data have;
set diagnoses;
if input(icd9code ,4.) >= 2910 and input(icd9code ,4.) <= 3458;
run;
Let me know in case of any queries.
If it is a character you cannot use the range. But you can use the in statement
SELECT * FROM s.diagnoses WHERE icd9code in ('2910','2911','2912');
To select range. You can define your own macro to generate strings of range like this
%macro range(start, stop);
%if &start. = &stop. %then %do;
"&stop."
%end;
%else %do;
"&start.", %range(%sysevalf(&start+1), &stop);
%end;
%mend range;
%put %range(2910, 2915);
* -> "2910", "2911", "2912", "2913", "2914", "2915"
Then assign it to a macro variable and use it in you where statement within proc sql
%let subset1 = %range(2910, 2915);
proc sql noprint;
create table want as
select *
from
have
where var_want in (&subset1.);
quit;
You can then define multiple subset variables with different ranges and combination them in where condition to achieve more complex subsetting.
For ranges you want to include in their entirety, you can use inequalities directly - no 'input' required, as long as you have leading zeros, and for the rest you can use in, e.g.
data example;
length char $1;
do i = 64 to 100;
char = byte(i);
output;
end;
run;
proc sql;
create table want as
select * from example where 'A' <= char <= 'Z' or char in ('[',']');
quit;
I have table lookup values as below
sno date
1 200101
2 200102
3 200103
4 200104
I wrote below macro
%let date=200102
proc sql;
select sno into :no from lookup where date=&date.;
quit;
I need a help on how to convert the entire table lookup into macro increment by creating first s.no and date as two macro variable then increment. So that i don’t need to update dates in my table lookup every time. So if i look up for date 201304 i need to get its corresponding s.no
Is there pattern to the SNO values? Are you basically numbering the months since 01JAN2001? If so then use INTCK() function.
data test;
input date yymmdd8. ;
format date yymmdd10. ;
sno = 1+intck('month','01JAN2001'd,date);
cards;
20010112
20010213
20010314
20010415
;
So you could create two macro variables. One with the base date and the other with the base SNO value.
36 %let basedate='01JAN2001'd ;
37 %let basesno=1;
38 %let date='01JAN2001'd ;
39 %let sno=%eval(&basesno + %sysfunc(intck(month,&basedate,&date)));
40 %put &=date &=sno;
DATE='01JAN2001'd SNO=1
41
42 %let date="%sysfunc(today(),date9)"d;
43 %let sno=%eval(&basesno + %sysfunc(intck(month,&basedate,&date)));
44 %put &=date &=sno;
DATE="16NOV2017"d SNO=203
If you want to simply translate one (unique) value into another. You can use (in)formats. They can do much more than just changing how data are read/displayed. They are easy to use, fast (in-memory) and don't depend on the table once created. Change the library to a permanent one if work (=> temporary library) doesn't suit your needs.
options fmtsearch=(formats,work);
data fmt(keep = fmtname type start end label hlo default);
length fmtname $10 type $1 start end $6 label 8 hlo $1 default 8;
fmtname = 'date_to_no';
type = 'I';
label=0;
do y = 2001 to 2099;
do m = 1 to 12;
start = put(y,4.) || put(m,z2.);
end = start;
label + 1;
default=50; /*default length of the string compared when informat is used. Should be higher than both start and end*/
output;
end;
end;
/*if you want to assign a value (=label) to inputs not found. In this case it's -2*/
hlo="O";
start = "";
end = start;
label= -2;
output;
run;
proc format library=work cntlin=fmt;
run;
data test;
no = input('200101',date_to_no.); output;
no = input('201710',date_to_no.); output;
no = input('201713',date_to_no.); output;
run;
Build a lookup table dynamically and create a macro variable for each row in the table. The macro variables will be named date_200101,date_200102,...and so on. They will contain a value equal to the corresponding sno value:
data lookup;
length var_name $20;
do sno = 1 to intck('month','01jan2001'd,date())+1;
date = input(put(intnx('month','01jan2001'd, sno-1, 'beginning'),yymmn6.),best.);
var_name = cats('date_',date);
call symput(var_name, cats(sno));
output;
end;
run;
You can then refer to the macro variables like so:
%let date =200103;
%put &&date_&date;
...or...
%put &date_200101;
The first usage example is using double macro resolution. Basically the macro processes needs to perform 2 iterations of the macro token &&date_&date in order to fully resolve it. On the first pass, it gets resolved to &date_200101. On the second pass, the macro token &date_200101 gets resolved to 1.
I have 169 towns for which I want to iterate a macro. I need the output files to be saved using the town-name (rather than a town-code). I have a dataset (TOWN) with town-code and town-name. Is it possible to have a %let statement that is set to the town-name for each iteration where i=town-code?
I know that I can list out the town-names using the index function, but I'd like a way to set the index function so that it sets a %let statement to the TOWN.town-name when i=TOWN.town-code.
All the answers below seem possible. I have used the %let = %scan( ,&i) option for now. A limitation is that the town names can be more than one word, so I've substituted underscores for spaces that I correct later.
This is my macro. I output proc report to excel for each of the 169 towns. I need the excel file to be saved as the name of the town and for the header to include the name of the town. Then, in excel, I merge all 169 worksheets into a single workbook.
%MACRO BY_YEAR;
%let townname=Andover Ansonia Ashford Avon ... Woodbury Woodstock;
%do i = 1999 %to 2006;
%do j = 1 %to 169;
%let name = %scan(&townname,&j);
ods tagsets.msoffice2k file="&ASR.\Town_Annual\&i.\&name..xls" style=minimal;
proc report data=ASR nofs nowd split='/';
where YR=&i and TWNRES=&j;
column CODNUM AGENUM SEX,(dths_sum asr_sum seasr_sum);
define CODNUM / group ;
define agenum / group ;
define sex / across ;
define dths_sum / analysis ;
define asr_sum / analysis ;
define seasr_sum / analysis ;
break after CODNUM / ul;
TITLE1 "&name Resident Age-Specific Mortality Rates by Sex, &i";
TITLE2 "per 100,000 population for selected causes of death";
run;
ods html close;
%end;
%end;
%MEND;
My guess is that the reason why you want to look up the town name by town index is to repeatedly call a macro with each town name. If this is the case, then you don't even need to get involved with the town index business at all. Just call the macro with each town name. There are many ways to do this. Here is one way using call execute().
data towns;
infile cards dlm=",";
input town :$char10. ##;
cards;
My Town,Your Town,His Town,Her Town
;
run;
%macro doTown(town=);
%put Town is &town..;
%mend doTown;
/* call the macro for each town */
data _null_;
set towns;
m = catx(town, '%doTown(town=', ')');
call execute(m);
run;
/* on log
Town is My Town.
Town is Your Town.
Town is His Town.
Town is Her Town.
*/
If you do need to do a table lookup, then one way is to convert your town names into a numeric format and write a simple macro to retrieve the name, given an index value. Something like:
data towns;
infile cards dlm=",";
input town :$char10. ##;
cards;
My Town,Your Town,His Town,Her Town
;
run;
/* make a numeric format */
data townfmt;
set towns end=end;
start = _n_;
rename town = label;
retain fmtname 'townfmt' type 'n';
run;
proc format cntlin=townfmt;
run;
%macro town(index);
%trim(%sysfunc(putn(&index,townfmt)))
%mend town;
%*-- check --*;
%put %town(1),%town(2),%town(3),%town(4);
/* on log
My Town,Your Town,His Town,Her Town
*/
Or how about you just pass both the code and the name to the macro as parameters? Like this?
%MACRO DOSTUFF(CODE=, NAME=);
DO STUFF...;
PROC EXPORT DATA=XYZ OUTFILE="&NAME."; RUN;
%MEND;
DATA _NULL_;
SET TOWNS;
CALL EXECUTE("%DOSTUFF(CODE=" || STRIP(CODE) || ", NAME=" || STRIP(NAME) || ");");
RUN;