I said: "Describe the data and describe what you would like to learn from this data, in relatively simple terms so that those of us not involved in this field of study can understand what the goals are. Do not try to describe the algorithm/mathematical/computational/statistical steps you might follow."   @Tom said: "In any case you need to ask for STATISTICAL help and not PROGRAMMING help. Once you know what analysis you need to perform you can ask for help on how to program it."   Please don't ignore these requests. This is, in my opinion, the only way to move forward. ... View more

I think Gremlins would be offended!   Just turn ODS GRAPHICS off at the start of your program if you don't want the graphics.   ods graphics off; ... View more

This is even easier than your solution   select * , case when "&value" = 'be' then '010' when "&value" = 'gc' then '001' else 'Other' end as newvar ... View more

@alepage wrote: No it seems that I need the single quote with a snowflake sql query. But it is a way to thanks that person for giving me the major part of the solution. Okay that's a good thing to do, but it would be even better if you explained to us why it was marked correct, without us having to ask. ... View more

Whenever SAS encounters a macro variable in code, it replaces the macro variable with its actual value, and then runs the code with these replacements made. So, for this code:   select * , case when &value. = 'be' then '010' when &value. = 'gc' then '001' End as newvar   This is the code that is actually executed:   select * , case when be = 'be' then '010' when be = 'gc' then '001' End as newvar   SQL thinks BE is the name of a variable in the table. (Do you understand why SAS thinks BE is the name of a variable in the table?) So this code only makes sense if the table that SQL is working on (TEMPTBL) has a variable named BE. Does it? A much better approach to create programs with macro variables is to get the code to work without macro variables first, and then substitute the macro variables into the code. But you haven't done that, and so it is highly recommended.   So, please tell us, what is the code you want to have executed here, without macro variables. Finally, when you are getting errors in the log, please show us (ALWAYS) the relevant parts of the log.     ... View more

I'm curious why you marked a solution correct when that solution didn't meet the requirements of having the date within single quotes. Is that no longer a requirement? ... View more

Use the round function with 2nd argument equal to 100 (which gives you answers 900 and 100) and then add 100.     ... View more

Digging into the problem as stated    I suspect there may be combinations of the outcome positive observations which may be significant for certain variables ...   Observations are never significant in the standard usage of the statistical term "significant". However, variables can be significant in a specified model or specified statistical test (which could be a simple t-test), but you don't ever in any statistical practice (that I am aware of) select certain observations that make the variable(s) significant. Furthermore if all you select are observations where the outcome variable is positive then no variable will ever be significant as there is no variability to predict or test.   May I make a suggestion? Describe the data and describe what you would like to learn from this data, in relatively simple terms so that those of us not involved in this field of study can understand what the goals are. Do not try to describe the algorithm/mathematical/computational/statistical steps you might follow. ... View more

I am struggling to understand the problem, and I really don't know what you are trying to do. It seems like you want to throw away some observations and keep others to achieve ... well, I don't really know but it sounds like you want to find observations to do a test ... what test? Shouldn't you keep all observations and do the test on all of them?    Maybe you could write a more complete and clear problem statement, stating clearly what you are planning to learn from this data and what tests you might consider in the analysis. ... View more

Perhaps an overview of the process for our self-admitted newbie @Sam_SAS might be helpful.   Arranging the data properly so you can actually perform the next analysis is a very important step, and one that beginners sometimes do not grasp. The best arrangement of the data is the "long" or "tall" data set as described by @Tom . Why is that the best? Because most SAS data analysis PROCs work best on data in this "long" or "tall" arrangement. So once you have this "long" or "tall" data set, a simple call to PROC FREQ (or other relevant PROC) gets the job done.   Now, the actual process of re-arranging such data isn't particularly simple, especially for newbies, and you have received several examples of working code. But this process of re-arranging the data into "long" or "tall" data sets is where you must start.   Note: there are of course examples where a "wide" dataset (rather than "long" or "tall") is required, but these are rare, and unless you have a solid justification for "wide" data sets, it is best to avoid them. See Maxim 19. Also note that when I talk about a "solid justification for wide data sets", this does not include situations where you want a wide output such as a wide report or wide Excel file; this is not a "solid justification for wide data sets". Wide outputs such as a wide report are wide Excel file are most easily produced from "tall" or "long" data sets via PROC REPORT or PROC TABULATE. ... View more

So it seems to me your real question is how to save the PROC IML results as actual SAS data sets. It would be desirable if you helped other users, and changed the title to reflect the real question you are asking.   IML has commands such as CREATE and APPEND to create actual SAS data sets from your results. Please see https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/imlug/imlug_langref_sect097.htm for many examples. ... View more

When you have lots of "good" and relatively few "bad", there is definitely the possibility that the variables you have does not predict the bads. In fact, this is a common situation. This is not necessarily your fault or the fault of the model, that's often the way it is. You can try oversampling (see here and here). You can also read a gazillion commentaries on oversampling, go to your favorite internet search engine and type in    statistical oversampling good vs bad ... View more

@SASGeek wrote: Any suggestions on how to output PROC REPORT to a specific cell? Perhaps you can fake this by having blank rows and columns in PROC REPORT (something I have never tried) so you could force the output to Excel to appear to begin in cell C6. But if there is already something else written to this Excel tab, even that won't work. ... View more

@SASGeek wrote: Any suggestions on how to output PROC REPORT to a specific cell?  This is not a capability of ODS EXCEL. Whatever you output via ODS Excel goes underneath  whatever has been previously output to that Excel worksheet (with a blank row in between the previous ODS Excel output and the new ODS Excel output). If nothing previously has been output to that Excel worksheet, then ODS Excel puts the output at the top left of the sheet. ... View more