I am not sure if UNICODE() could work. proc iml; x = {0.1 0.2, 0.01 0.05, 0.03 0.12}; c = {"\u8023" "\u8801"}; cc=unicode(c); print cc; print x[colname=cc]; quit; ... View more

Did you ever figure out a solution to this issue? I am facing the same situation. I have close to 200 variables that I need to impute, most of which are ordinal. I am using FCS since it can handle continuous, ordinal, and discrete data. However, the full imputation model takes about three days to run. OUCH! Moreover, I have too many variables for my sample size (498). Consequently, the model was not able to impute several variables; basically, it could not impute the majority of continuous variables that also happened to have the highest missing response rate. Hence, I would like to break up the imputation in two parts: use MCMC for the continuous data and FCS for the ordinal and discriminant data. Unfortunately, I cannot see how to exclude variables from being imputed but still used them as predictors. How did you resolve this problem? There is no real advantage to using the method that was suggested to your question because it would not save runtime.   Cristian ... View more

okay, that was easier than I thought. I got everything that I need.   Thank you everyone.   Cristian ... View more

crisgugiu, The macro store is the most effective way to store your code and hide the code.  However, if you want to save the code, you can do what I do so when I ship updates, I send over one sas7bcat file isntead of a zipped file of the code. There are two ways to store SAS code to a catalog.  You can either store each data step in a SAS catalog (big pain), or you can load an entire program to a SAS catalog.  Unlike storing SAS Macros to the macro catalog where the Macro catalog is set as a libref, set the catalog as a fileref (i.e., filename fileref "path").  There are two ways to do this: Write the program to the file: libname catlib "path-to-catalog"; filename catfile catalog "catlib.project"; data _null_;    file catfile(filename.source);     put "data newdata;";     put "    set olddata;";     put "  * some more datasteps;";     put "run;"; run; But, since I'm lazy and it's not as simple to test programs that way and make adjustments: libname catlib "path-to-catalog"; filename catfile catalog "catlib.project"; filename includes "my include library"; /* it's already set up, but for clarity, I've included it here */ data _null_;     file catfile(filename.source);    /* note the ".source" - need that so it can be used in an include */     infile includes(filename.sas);  /* note - the ".sas" is needed where the OS requires it to work */     input;     put _infile_; run; With the second step, filename should be the same for both the input and output so when you include it in code later on, your code doesn't break. So you know, I get a listing of all the files in the include library first, then use a macro to loop through the list and build my source catalog.  Later, when you use the catalog, do the following: libname catlib "path-to-catalog"; filename catfile catalog "catlib.project"; %include catfile(filename); You can prevent the actual source code from being viewed in the process by creating a macro that stores the current options, sets options nosource nosource2;, then re-stores the options to what they were.  If you want to go to the trouble. # ... View more

Is there are way to attach a SAS data set? Here is the SAS code. ods graphics on; proc cluster data=HCA_std method=WARD TRIM=10 NONORM k=10 ccc pseudo print=30 outtree=Tree ;    var can:;    id NationYR;    copy FH Van Democ Elecsd Polity;    format NationYR $12.; run; ods graphics off; goptions vsize=10in hsize=10in htext=.5pct htitle=2pct ftext='Garamond' ; title "Cluster Analysis of Country/Election Years"; axis1 order=(0 to 1 by 0.1) label = (height=2pct) ; proc tree data=Tree out=Clusters(drop=CLUSNAME can:) nclusters=2 similar           haxis=axis1 VPAGES=2 horizontal  ; where _FREQ_ >-1; height _rsq_; copy FH Van Democ Elecsd Polity; id NationYR; run; ... View more

Hey Cristian, The number of band colors and their order can be control via the goptions COLORS=() option. You may be able to set the number of colors in a way to match your ranges. There is currently not a way to assign ranges to specific colors. Hope this helps, Dan ... View more