i have a data that contain 30 variable and 2000 Observations.
I want to calculate regression in a loop, whan in each step I delete the i row in the data.
so in the end I need thet my output will be 2001 regrsion, one for the regrsion on all the data end 2000 on each time thet I drop a row.
I am new to sas, and I tray to find how to do it withe macro, but I didn't understand.
Any comments and help will be appreciated!
This will create the data set I was talking about in my comment to Chris.
data del1V /view=del1v;
length group _obs_ 8;
set sashelp.class nobs=nobs;
_obs_ = _n_;
group=0;
output;
do group=1 to nobs;
if group eq _n_ then;
else output;
end;
run;
proc sort out=analysis;
by group;
run;
DATA NEW;
DATA OLD;
do i = 1 to 2001;
IF _N_ ^= i THEN group=i;
else group=.;
output;
end;
proc sort data=new;
by group;
proc reg syntax;
by group;
run;
This will create a data set that is much longer. You will only call proc reg once, but it will run 2001 models.
Examining 2001 regression outputs will be difficult just written as output. You will likely need to go read the PROC REG support documentation and look into the output options for whatever type of output you're interested in. SAS can create a data set with the GROUP column to differentiate the results.
I edited my original answer per #data null suggestion. I agree that the above is probably faster, though I'm not as confident that it would be 100x faster. I do not know enough about the costs of the overhead of proc reg versus the cost of the group by statement and a larger data set. Regardless the answer above is simpler programming. Here is my original answer/alternate approach.
You can do this within a macro program. It will have this general structure:
%macro regress;
%do i=1 %to 2001;
DATA NEW;
DATA OLD;
IF _N_=&I THEN DELETE;
RUN;
proc reg syntax;
run;
%end;
%mend;
%regress
Macros are an advanced programming function in SAS. The macro program is required in order to do a loop of proc reg. The %'s are indicative of macro functions. &i is a macro variable (& is the prefix of a macro variable that is being called). The macro is created in a block that starts and ends with %macro / %mend, and called by %regress.
Examining 2001 regression outputs will be difficult just written as output. You will likely need to go read the PROC REG support documentation and look into the output options for whatever type of output you're interested in. Use &i to create a different data set each time and then append together as part of the macro loop.
Related
I have two values which represent dates:
a=101 and b=103
Below is first macro saved in separate file one.sas:
%global time nmall;
%let nmall =;
%macro pmall;
%do i=&a. %to &b;
%if &i =&a. then %do;
%let nmall=&nmall.&i;
%end;
%else %let nmall=&nmall.,&i;
end;
%put (&nmall);
%mend;
%pmall;
So above pmall give me values 101,102,103.
Below is second macro:
%include “one.as”;
%macro c(a=,b=);
%let m=;
%let m1=;
%do i =&a %to &b;
%let o=&i;
proc sql;
create table new&o as select * from data where nb in(&o.);quit;
%let m =&m.date&o;
data date&o.;
set date&o.;
if pass =&o.;
run;
proc sort data=date&o.;
by flag;
end;
data output &a._&b.;
set &m;
%mend;
The above macro creates three datasets date101 date102 and date 103, then append it to output101_103.
I am trying to modify above macros in such a way that I will not use %macro and %mend approach. Below is the modified macro code:
data a_to_c;
do o=&a to &c;
output;
end;
run;
so above code will have values 101 102 103 in variable o for dataset a_to_c.
data _null_;
set a_to_c;
call execute ('create table new’||strip(o)||' as select * from data
where nb in(’||strip(o)||' );quit;’);
run;
I want to know how to do below things.
Create pmall values in a macro variable in my modified macro inside the data step data a_to_c, so that I can use it further.
How to proceed from %let m macro in the first macro code to new code which I am developing above.
Geetha:
I think you will find the macro-ization of the process to be far easier if you go from a data-centric explicit solution and proceed abstracting the salient features into macro symbols (aka variables)
The end run solution appears to be:
data output_101_to_103;
set original_data;
where nb between 101 and 103;
run;
proc sort data=output_101_to_103;
by nb flag;
run;
In which case you could code a macro that abstracts 101 to FIRST and 103 to LAST. The data sets could also be abstracted. The abstracted parts are specified as the macro parameters.
%macro subsetter(DATA=, FIRST=, LAST, OUTPREFIX=OUTPUT);
%local out;
%let out = &OUTPREFIX._&FIRST._&LAST.;
data &out;
set &DATA.;
where nb between &FIRST. and &LAST.;
* condition = "between &FIRST. and &LAST."; * uncomment if you want to carry along the condition into your output data set;
run;
proc sort data=&out;
by nb flag;
run;
%mend;
And use as
%subsetter (data=original_data, first=101, last=103, outprefix=output)
Note: If you did keep the condition variable in the output data, you WOULD NOT be able to use it directly as a source code statement in a future data step, as in if nb condition then ...
I suppose you could also pass the NB and FLAG as parameters -- but you approach a point of diminishing returns on the utility of the macro.
Macro-izing the specific example I showed doesn't make too much sense unless you need to perform a lot of different variations of FIRST and LAST in a well documented framework. Sometimes it is just better to not abstract the code and work with the specific cases. Why? Because when there are too many abstracted pieces the macro invocation is almost as long as the specific code you are generating and the abstraction just gets in the way of understanding.
If the macro is simply chopping up data and reassembling data, you might be better served rethinking the flow using where, by, and class statements and abstracting around that.
Pmall is macro variable which will have list of values separated by
commas. In my modify macro, i want to create pmall as macro variable
in the datastep data a_to_c; do o=&a to &c; output; end; run; – geetha
anand 1 min ago
To create a macro variable from within a data step using the CALL SYMPUTX() function.
data a_to_c;
length pmall $200 ;
do o=&a to &c;
pmall=catx(',',pmall,o);
output;
end;
call symputx('pmall',pmall);
drop pmall;
run;
If you really want to generate code without a SAS macro you can use CALL EXECUTE() or write the code to a file and use %INCLUDE to run it. Or for small pieces of code you could try putting the code in a macro variable, but macro variables can only contain 64K bytes.
It is really hard to tell from what you posted what code you want to generate. Let's assume that you want to generate an new dataset for each value in the sequence and then append that to some aggregate dataset. So for the first pass through the loop your code might be as simple as these two steps. First to create the proper subset in the right order and the second to append the result to the aggregate dataset.
proc sort data=nb out=date101 ;
where nb=101 ;
by flag ;
run;
proc append base=date101_103 data=date101 force;
run;
Then next two times through the loop will look the same only the "101" will be replaced by the current value in the sequence.
So using CALL EXECUTE your program might look like:
%let a=101;
%let c=103;
proc delete data=date&a._&c ;
run;
data _null_;
do nb=&a to &c;
call execute(catx(' ','proc sort data=nb out=',cats('date',nb,'),';'));
call execute(cats('where nb=',nb,';')) ;
call execute('by flag; run;');
call execute("proc append base=date&a._&c data=");
call execute(cats('date',nb));
call execute(' force; run;');
end;
run;
Writing it to a file to run via %INCLUDE would look like this:
filename code temp ;
data _null_;
file code ;
do nb=&a to &c;
put 'proc sort data=nb out=date' nb ';'
/ ' where ' nb= ';'
/ ' by flag;'
/ ';'
/ "proc append base=date&a._&c data=date" nb 'force;'
/ 'run;'
;
end;
run;
proc delete data=date&a._&c ;
run;
%include code / source2;
If the goal is to just create the aggregate dataset and you do not need to keep the smaller intermediate datasets then you could just use the same name for the intermediate dataset on each pass through the loop. That will make the code generation easier as then there is only only place that needs to change based on the current value. Also that way you only need to have two dataset names even for a sequence of 10 or 20 values. It will take less space and reduce clutter in the work library.
I'm looking for a way to use a series of values for a macro parameter instead of a single value. I'm basically manipulating a series of files for consecutive months (May 2014 to Sept 2015) and I've written a macro to take advantage of the naming conventions. However, I'm still manually writing out the months to use the macro. I'm doing this many times over with lots of different files from this month. Is there a way to have the parameter reference a list of values and go through them like an array/do-loop? I've looked into %ARRAY as a possibility but that doesn't seem to do what I'm looking for unless I'm not seeing it's full capability. I've attached a sample of this code below.
%MACRO memmonth(monyr=);
proc freq data=work.both_&monyr ;
table var1/ out=work.mem_&monyr;
run;
data work.mem_&monyr;
set work.mem_&monyr;
count_&monyr=count;
run;
%MEND memmonth;
%memmonth(monyr=may14)
%memmonth(monyr=jun14)
%memmonth(monyr=jul14)
%memmonth(monyr=aug14)
%memmonth(monyr=sep14)
%memmonth(monyr=oct14)
%memmonth(monyr=nov14)
%memmonth(monyr=dec14)
%memmonth(monyr=jan15)
%memmonth(monyr=feb15)
%memmonth(monyr=mar15)
%memmonth(monyr=apr15)
%memmonth(monyr=may15)
%memmonth(monyr=jun15)
%memmonth(monyr=jul15)
%memmonth(monyr=aug15)
%memmonth(monyr=sep15)
In general I would recommend passing the list of values as a space delimited list and adding looping logic in the macro. If spaces are valid characters in the values then use some other delimiter. Do not use comma as the delimiter as it means you will need to use macro quoting to call the macro.
So your basic macro is this.
%macro memmonth(monyr);
proc freq data=work.both_&monyr ;
table var1/ out=work.mem_&monyr (rename=(count=count_&monyr)) ;
run;
%mend memmonth;
%memmonth(may14)
%memmonth(jun14)
You could change it to this.
%macro memmonth(monyrlist);
%local i monyr;
%do i=1 %to %sysfunc(countw(&monyrlist));
%let monyr=%scan(&monyrlist,&i);
proc freq data=work.both_&monyr ;
table var1/ out=work.mem_&monyr (rename=(count=count_&monyr)) ;
run;
%end;
%mend memmonth;
%memmonth(may14 jun14)
If you always want to process all of the months in an interval then you could just pass in the start and end month of the interval.
%macro memmonth(start,end);
%local i monyr;
%do i=0 %to %sysfunc(intck(month,"01&start"d,"01&end"d));
%let monyr=%sysfunc(intnx(month,"01&start"d,&i),monyy5.);
proc freq data=work.both_&monyr ;
table var1/ out=work.mem_&monyr (rename=(count=count_&monyr)) ;
run;
%end;
%mend memmonth;
%memmonth(may14,sep15)
If you have a source list, whether it is a text file or a dataset, you can use a simple data step to generate macro calls. So if you have an input dataset with the variable MONYR then your driver program would look like this:
data _null_;
set mylist ;
call execute(cats('%nrstr(memmonth)(',MONYR,')'));
run;
If the source is a file with the names then replace the SET statement with the appropriate INFILE and INPUT statements. If the source is a directory name then look at one of the many SAS ways to read the names of files in a directory into a dataset and use that to drive the macro call generation.
In SAS, how can I assign a variable coming from either the OUTEST or OUTSTAT functions to be used in a loop?
For example, say I want to run some sort of iterative analysis until my mean (average) reaches a certain threshold. I know how to extract the mean using either OUTEST or OUTSTAT, but then how can I perform operations or blocks of code on it?
Thank you.
If you are interested in details, I am trying to perform backward selection of VIFs (to remove multicollinearity). Unfortunately, SAS doesn't seem to have a 'SELECTION=BACKWARD' feature for this...
EDIT: Updated with sample code:
%MACRO MULTICOLLINEARITY(TABLE_SUFFIX,YVAR,FIELDS,MAX_VIF);
/* PRELIMINARY PROC REG ON ALL FIELDS*/
PROC REG DATA=TABLE_&TABLE_SUFFIX. NOPRINT;
MODEL &YVAR = &FIELDS / VIF COLLIN NOINT;
ODS OUTPUT PARAMETERESTIMATES=PAREST1;
RUN;
/* RETAIN NON-NULL VIF FIELDS ONLY */
DATA NO_NULL_VIF;
SET PAREST1 (WHERE=(VarianceInflation <> .));
RUN;
/* CREATE VARIABLE LIST OF NON-NULL VIF FIELDS */
PROC SQL;
SELECT VARIABLE
INTO :NO_NULL_VIF_FIELDS SEPARATED BY ' '
FROM NO_NULL_VIF;
QUIT;
/* RE-RUN REGRESSION WITH NON-NULL VIF FIELDS ONLY */
PROC REG DATA=TABLE_&TABLE_SUFFIX. NOPRINT;
MODEL &YVAR = &NO_NULL_VIF_FIELDS / VIF COLLIN NOINT;
ODS OUTPUT PARAMETERESTIMATES=PAREST2;
RUN;
/* START ITERATION OF DROPPING THE HIGHEST VIF UNTIL THE CRITERIA IS MET */
???
%MEND;
%MULTICOLLINEARITY(, RESPONSE, &INPUT_FIELDS,???)
And by criteria I mean VIF_MAX < N where N is some threshold specified in the macro. For example, if we want to retain only fields with VIF less than 5, then it should drop the highest one, re-run the PROC REG, drop the highest, re-run, etc. etc. until the highest on is less than 5.
First off - I'd verify that you can't do this using PROC MODEL. I'm not a regression guy so I don't know for sure. Might be worth posting on a more stat-focused site; CV isn't really appropriate since they're not generally trying to answer software questions, but maybe communities.sas.com . I would find it surprising if this wasn't directly possible in PROC MODEL and/or in one of the more complicated procs.
Second, the way I'd approach this is to write a recursive macro. Take out the first part (the non-null VIF fields) and either move that to an outer macro that just runs once, or make it an expectation of the programmer to do on his/her own (unless this is not feasible, and/or can change with iterations - not something I'm knowledgeable of). Then do something like this:
%MACRO MULTICOLLINEARITY(TABLE_SUFFIX,YVAR,FIELDS,MAX_VIF);
ods _all_ close;
%put Running with &fields; *note which fields currently running;
*also may want to include a run # counter as parameter;
PROC REG DATA=TABLE_&TABLE_SUFFIX.;
MODEL &YVAR = &FIELDS / VIF COLLIN NOINT;
ODS OUTPUT PARAMETERESTIMATES=PAREST2;
RUN;
quit;
*Data step to analyse PAREST2 and see if any of the fields can be dropped;
proc sort data=parest2;
by descending varianceinflation;
run;
data _null_;
set parest2(obs=1);
if varianceinflation > &max_vif then do;
fields_run = tranwrd("&fields",trim(variable),' ');
if not missing(fields_run) then do;
call_string = cats('%multicollinearity(',"&table_suffix.,&yvar.,",fields_run,",&max_vif.)");
call execute(call_string);
end;
end;
else do;
put "Stopped with Max VIF:" variable "=" varianceinflation;
run;
ods preferences;
%MEND MULTICOLLINEARITY;
Then you call it once with the full field list, and it calls itself in the CALL EXECUTE if there is still a parameter left. An incremented # of runs may be helpful (both to see how many times it ran in your log, and to be able to make sure that you don't end up in an infinite loop if you make a mistake with the fields variable deletion.)
I would run this with OPTION NONOTES NOSOURCE; and none of the symbogen/mprint stuff on, so you can just get the %put/put statements in your log.
I am trying to determine the number of observations in a dataset, then convert this number into a macro variable that i can use as part of a loop. I've searched the web for answers and not had much luck. I would post some example code I've tried but I have literally no idea how to approach this.
Could anybody assist?
Thanks
Chris
SAS stores dataset information, such as number of observations, separately, so the key is to access this information without having to read in the entire dataset.
The following code will do just that, the if 0 part is never true so the dataset isn't read, however the information is.
data _null_;
if 0 then set sashelp.class nobs=n;
call symput('numobs',n);
stop;
run;
%put n=&numobs;
You can also get it from dictionary.tables like this:
proc sql noprint;
select nobs into :nobs
from dictionary.tables
where libname='YourLibrary' and memname='YourDatasetName';
quit;
Here it is:
Create macro variable:
data _null_;
set sashelp.class;
call symput("nbobs",_N_);
run;
See result:
%put &nbobs;
Use it:
data test;
do i = 1 to &nbobs;
put i;
end;
run;
I want to perform some regression and i would like to count the number of nonmissing observation for each variable. But i don't know yet which variable i will use. I've come up with the following solution which does not work. Any help?
Here basically I put each one of my explanatory variable in variable. For example
var1 var 2 -> w1 = var1, w2= var2. Notice that i don't know how many variable i have in advance so i leave room for ten variables.
Then store the potential variable using symput.
data _null_;
cntw=countw(¶meters);
i = 1;
array w{10} $15.;
do while(i <= cntw);
w[i]= scan((¶meters"),i, ' ');
i = i +1;
end;
/* store a variable globally*/
do j=1 to 10;
call symput("explanVar"||left(put(j,3.)), w(j));
end;
run;
My next step is to perform a proc sql using the variable i've stored. It does not work as
if I have less than 10 variables.
proc sql;
select count(&explanVar1), count(&explanVar2),
count(&explanVar3), count(&explanVar4),
count(&explanVar5), count(&explanVar6),
count(&explanVar7), count(&explanVar8),
count(&explanVar9), count(&explanVar10)
from estimation
;quit;
Can this code work with less than 10 variables?
You haven't provided the full context for this project, so it's unclear if this will work for you - but I think this is what I'd do.
First off, you're in SAS, use SAS where it's best - counting things. Instead of the PROC SQL and the data step, use PROC MEANS:
proc means data=estimation n;
var ¶meters.;
run;
That, without any extra work, gets you the number of nonmissing values for all of your variables in one nice table.
Secondly, if there is a reason to do the PROC SQL, it's probably a bit more logical to structure it this way.
proc sql;
select
%do i = 1 %to %sysfunc(countw(¶meters.));
count(%scan(¶meters.,&i.) ) as Parameter_&i., /* or could reuse the %scan result to name this better*/
%end; count(1) as Total_Obs
from estimation;
quit;
The final Total Obs column is useful to simplify the code (dealing with the extra comma is mildly annoying). You could also put it at the start and prepend the commas.
You finally could also drive this from a dataset rather than a macro variable. I like that better, in general, as it's easier to deal with in a lot of ways. If your parameter list is in a data set somewhere (one parameter per row, in the dataset "Parameters", with "var" as the name of the column containing the parameter), you could do
proc sql;
select cats('%countme(var=',var,')') into :countlist separated by ','
from parameters;
quit;
%macro countme(var=);
count(&var.) as &var._count
%mend countme;
proc sql;
select &countlist from estimation;
quit;
This I like the best, as it is the simplest code and is very easy to modify. You could even drive it from a contents of estimation, if it's easy to determine what your potential parameters might be from that (or from dictionary.columns).
I'm not sure about your SAS macro, but the SQL query will work with these two notes:
1) If you don't follow your COUNT() functions with an identifier such as "COUNT() AS VAR1", your results will not have field headings. If that's ok with you, then you may not need to worry about it. But if you export the data, it will be helpful for you if you name them by adding "...AS "MY_NAME".
2) For observations with fewer than 10 variables, the query will return NULL values. So don't worry about not getting all of the results with what you have, because as long as the table you're querying has space for 10 variables (10 separate fields), you will get data back.