Did any one use Macro Pairwise_chisq before? Some output files like ChiTest..have no obs...

[xzhang]

Hello

I used Pairwise_chisq, but some output files like Chitest have no observations. I tried many times, but it did not work. Did any one use it before?

I referred to this paper: http://support.sas.com/resources/papers/proceedings14/1544-2014.pdf.

Thank you.

[Reeza]

If you post your full code with some sample data and how you called it I'll take a look at it. 

---

@xzhang wrote:

Hello

I used Pairwise_chisq, but some output files like Chitest have no observations. I tried many times, but it did not work. Did any one use it before?

I referred to this paper: http://support.sas.com/resources/papers/proceedings14/1544-2014.pdf.

Thank you.

---

[ballardw]

---

@xzhang wrote:

Hello

I used Pairwise_chisq, but some output files like Chitest have no observations. I tried many times, but it did not work. Did any one use it before?

I referred to this paper: http://support.sas.com/resources/papers/proceedings14/1544-2014.pdf.

Thank you.

---

Doesn't work is awful vague.

Are there errors in the log?: Post the code and log in a code box opened with the {i} to maintain formatting of error messages.

No output? Post any log in a code box.

Unexpected output? Provide input data in the form of data step code pasted into a code box, the actual results and the expected results. Instructions here: https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-dat... will show how to turn an existing SAS data set into data step code that can be pasted into a forum code box using the {i} icon or attached as text to show exactly what you have and that we can test code against.

[xzhang]

Thank you for reply. Here are the codes

proc freq data=a;

tables b*c/chisq nocol norow nopercent out=d;

run;

proc contents data=d;

run;

　

data d; set d;

b=STRIP(put(b_c,8.));

c=STRIP(put(c_c,8.));

run;

proc contents data=d;

run;

%macro Pairwise_Chisq(data=, group=, outcome=, count=);

proc freq data=&data noprint;

table &group/out=out_group;

run;

data groupID;

set out_group;

GID=_n_;

run;

data _null_;

set groupID nobs=nobs;

call symput('num_group', compress(put(nobs, 11.)));

stop;

run;

data ChiTest; *store pairwise Chisquare statistics to this;

stop;

run;

%Do group1=1 %to &num_group-1; *for each pair obtain chisquare stat;

%Do group2=&group1+1 %to &num_group;

data group_pair;

set groupID;

if GID EQ &group1 then call symput('name_group1', type);

if GID EQ &group2 then call symput('name_group2', type);

run;

proc freq data=&data noprint; output PCHI out=CTout ;

tables &group*&outcome /chisq cellchi2 nopercent;

where &group in ("&name_group1", "&name_group2");

weight &count;

run;

data CTout1;

set CTout;

PAIR_COMPARED="&name_group1"||" vs "||"&name_group2";

run;

data ChiTest;

set ChiTest CTout1;

run;

%End;

%End;

data ChiTest; *rename variables for final results;

set ChiTest;

CHI_STAT=_PCHI_; DF=DF_PCHI; P_VALUE=P_PCHI;

keep pair_compared Chi_stat DF p_value;

run;

%mend Pairwise_Chisq;

%Pairwise_Chisq(data=d,group=b_c,outcome=c_c,count=count);

_________________________________________________________________

no output, only output 5 sas files.

[xzhang]

there were something I edited wrong. Here is the right one.

data d; set d;

b_c=STRIP(put(b,8.));

c_c=STRIP(put(c,8.));

run;

[Reeza]

Where does A come from? Sample data please.

[xzhang]

A is raw data, b (3 levels) and c (3 levels) are two categorical variables. like:

Id  b  c

1   15  0

2   16   1

3    17  2

4     16  0

..............

[Reeza]

I need data to run it with. 4 lines won't be enough. 

[ballardw]

One potential issue is the use of comments like:

*store pairwise Chisquare statistics to this;

Inside a macro the comments should

%*store pairwise Chisquare statistics to this;

or

/*store pairwise Chisquare statistics to this*/

This data step has issues because the data set groupid has no variable named type.

data work.group_pair;
set work.groupID;
if GID EQ &group1 then call symput('name_group1', type);
if GID EQ &group2 then call symput('name_group2', type);
 
run;

so the values for the name_group macro variables are blank. And therefore the proc freq fails as no values of the group variable will match the blanks in &name_group1 and 1.

proc freq data=&data noprint;
output PCHI out=work.CTout ;
tables &group*&outcome /chisq cellchi2 nopercent;
where &group in ("&name_group1", "&name_group2");
 
weight &count;
run;

[xzhang]

Thank you sooooooooooo..much! I changed the name "type" to the corresponding variable name in my data. It is working now. Much appreciated!