I'm sure that there are more technically correct ways to do this, but one really easy solution is to append an invisible character onto the date. You can type invisible characters into the program editor by turning on Num Lock, holding down the Alt key, typing the number 255, then releasing the Alt key. Try running the following program. Note that you will need to recreate the invisible character in your SAS session before you submit the program. Both dates look the same the first time Excel is opened. If you make a change, save, and reopen excel, however, the value for date1 changes and the value for date2 does not. Hope this helps. [pre] data temp; date1 = '01may2009'd; date2 = cats(substr(propcase(put(date1, monyy5.)), 1, 3),'-',substr(put(date1, monyy5.),4,2), ' '); /*The last argument is an invisible character created with Alt+255*/ format date1 monyy7.; run; ods listing close; ods html file='C:\temp.xls'; proc print data=temp; run; ods html close; [/pre] ... View more

Enter the following statement at the top of the config file, substituting for the desired directory: - SASINITIALFOLDER = 'directory' ... View more

Never mind, I figured it out by modifying the config file. Thanks again. ... View more

In addition to the intnx function, you could also use the weekday function. For example: %let weekday=%sysfunc(weekday(%sysfunc(today()))); If you create the macro variable weekday as above, it will have a value of 2 every Monday. Conditional logic could then be used to determine which date variables to use. ... View more

How are you determining that the "length" of B is four? This is just a guess, but maybe you successfully changed the length of the variable but left the format unchanged. Try adding a couple of format statements to the code and see if anything changes. For example: data comb; set one_n two_n; format A B $5.; run; ... View more

The libref function returns a zero (not a one) if a libref is found. For example, try the following: %macro test (lib=SASHELP); %if (%sysfunc(LIBREF(&lib))=0) %then %put Library Assigned; %mend; %test; I'm not sure what the existance of a library has to do with the sysmsg function. The sysmsg function might return a missing value even if the library has been assigned. ... View more

[pre] data test; input phyid flag $; datalines; 1001 C 1002 P 1003 C 1003 P 1004 C 1004 P ; run; proc sql; select phyid from test where flag = 'C' intersect select phyid from test where flag = 'P'; quit; [/pre] ... View more

Since you are looking for a Wednesday, you could also use 'middle' as the alignment parameter instead of incrementing by three. data _null_; LastWednesday = intnx('week', today(), 0, 'middle'); if LastWednesday >= today() then LastWednesday + -7; put LastWednesday = weekdate.; run; ... View more

Ok, now I see. Without the %*; sequence, the tab which precedes the %unquote is passed to the input stack. Thanks! Message was edited by: polingjw ... View more

Chang, What does %*; do? The sequence appears twice in your solution. I can't remember seeing this before, so I'm just curious. Thanks. ... View more

You need to remove the two leading ampersands from the three global statements in the beginning of the macro. For example, when &i resolves to 1, the statement %global &&skodf&i; will resolve to %global &skodf1;, which produces an error because a macro variable name cannot include an ampersand. This explains the error messages you referenced. I suspect, however, that there may be a couple of additional problems with this code that you will still need to resolve. ... View more

Here is an alternative solution. This solution uses one sql step and then a single pass through the data with a data step: [pre] %macro recode(dataset, vars); proc sql noprint; %let i=1; %do %until(%scan(&vars, &i) = ); %let var = %scan(&vars, &i); %local &var; select &var into: &var separated by ' ' from (select &var, count(*) as count from &dataset where &var ne '' group by &var) order by count, &var desc; %let i = %eval(&i + 1); %end; quit; data &dataset._recode; set &dataset; %let i = 1; %do %until(%scan(&vars, &i)= ); %let var = %scan(&vars, &i); %let j=1; %do %until(%scan(&&&var, &j)= ); %if %eval(&j > 1) %then else ; if &var = "%scan(&&&var, &j)" and &var ne "" then &var._recode = &j; %let j = %eval(&j + 1); %end; else &var._recode = .; %let i = %eval(&i + 1); %end; run; %mend; data one; input id (v1-v5) ($); cards; 1 a a a . a 2 a b b . a 3 a c c . a 4 b c d . a 5 b c e . a 6 c c . . a ; run; %recode(one, v1 v2 v3 v4 v5) proc print data=one_recode; run; [/pre] Message was edited by: polingjw Message was edited by: polingjw ... View more

You need to create a data set with the values of the independent variables that you want to make predictions for, concatenate this data with the original data set (leaving the value of the response variable as missing), then rerun the regression model specifying the P option in the model statement. ... View more

Instead of using the NLDATM function, would something like this work? data _null_; worddate = 'Wed Sep 15 12:28:27 2010'; numberdate = input(scan(worddate, 3, ' ')!!scan(worddate, 2, ' ')!!scan(worddate, 5, ' ')!!':'!!scan(worddate, 4, ' '), datetime18.); put 'The SAS date ' numberdate 'is ' worddate; run; ... View more