@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

ODS Excel does not let you place two graphs side-by-side. It can only place two plots one above the other.   You could get SAS two place two graphs side-by-side and then output this to an Excel sheet using ODS Excel, so they appear in Excel side by side. I have used SAS to put two graphs side by side, but I have not used ODS EXCEL to get the two graphs into Excel. See: https://blogs.sas.com/content/iml/2022/07/13/ods-graphs-panel.html ... View more

Get the error messages saying operator requires compatible formats - makes sense the variable "sampdate" is formatted MMDDYY10.   How can I format the variable QSTART as MMDDYY10. - either inside or outside a macro?  Ideally, I'd like to create a correctly formatted global "QSTART" variable before any of the programs run.   I don't think the error message was referring to formats. (Which is why you should always quote the exact error message and always show us the log). It was referring to numeric vs character. You were trying to compare a numeric variable to a character variable, and SAS doesn't know how to do that. Regarding formats, it is possible and even easy to compare date values in format MMDDYY10. to date values in format DATE9. for example, SAS doesn't use the formatted values in its comparisons, it always uses the unformatted values to do arithmetic or logical comparisons.   Anyway, as @Kurt_Bremser says, you can use unformatted date values, that works when you do it properly; or as @Tom says, you can use date literals such as '01JAN2020'd, or with a macro variable it would be "&qstart"d where &qstart must have a value that is a proper date literal inside the quotes. ... View more

@LinusH wrote: Your example is abstract, but even so your "want" data set looks hard to use. What is the purpose of the different variables? Without knowing the application I would suggest transposing the data instead. This removes all blank cells, and keeps the lineage from which columns the values originated. @LinusH makes a very good point, that sometimes people spend time and effort converting a data set layout that is difficult to use to another data set layout that is equally difficult to use. It is time and effort for no benefit.   So instead of telling us what the re-arrangement of the data is desired, @Lost_Gary should tell us the final result of this programming effort, what tables/plots/analysis are you going to wind up doing with this data. Then we can advise in a much more specific fashion. ... View more

I often have to analyze all the data from all years at once, and I am wondering if it is better practice to create dummy variables for all variables across all years or to just maintain the variables used for that particular year. Not sure what you mean by "dummy variables", but I would have all the variables in the data set, and missing values when the question isn't asked. The phrase "dummy variables", as I am understanding its meaning, is usually used to indicate a binary (or 0/1) variable, and that doesn't make any sense to do here when the value is actually missing.   As an example, the survey has a language section. In year 2018, there was a variable for lang_chinese (0 or 1) whereas in 2024 there are options for lang_mandarin (0 or 1) and lang_cantonese (0 or 1) and lang_chinese was removed. So is it better to create dummy variables for all 3 across years or not? There is no way to answer this, as we don't know what analysis you intend to do. If we did know what analysis you are going to do (hint hint hint), we could provide some possible suggestions.   I also comment on your usage of terminology ... you say "there was a variable for lang_chinese" but the variable is really "Speaks Chinese". Please do not use lang_chinese (the name of the variable) to indicate the meaning of the variable, which is "Speaks Chinese". You want to create a label for this variable, so in your reports and plots the words "Speaks Chinese" appear rather than the variable name lang_chinese. The variable name and the variable meaning should not be used interchangeably.   ... View more

@sbxkoenk is correct, this is a numerical precision issue. Here is more reading: Solved: comparing two values - SAS Support Communities and even more at https://communities.sas.com/t5/SAS-Programming/Formula-evaluates-totals-the-same-set-of-variables-differently/m-p/761364#M240925   Your value of Z is not -90 because SAS cannot represent some decimals exactly. Decimal values that are not an integer power of 2 cannot be represented exactly. So y=3.6 produces a number that isn't quite 3.6 (can't represent it exactly) and so the value of Z (which uses Y) isn't exactly -90.    If you display the numbers using the HEX format so we can see the exact value, like this   data three; x=36; y=3.6; z=(y-x)/x*100; minusninety=-90; if z<=-90 then a=1; else a=0; if a=-90 then flag=1; run; proc print data=three; var x y z minusninety; format x y z minusninety hex16.; run;   this is the result, note that variable Z is not exactly equal to the value of variable MINUSNINETY. So then variable A=0 because it is not exactly equal.    Obs x y z minusninety 1 4042000000000000 400CCCCCCCCCCCCD C0567FFFFFFFFFFF C056800000000000   Then flag is missing because it only is assigned a value when A=1, but A=0 so no value is assigned.   I think the best solution is to use the ROUND function like this: if round(z)=-90 as suggest by @Kurt_Bremser here is a better solution which will work in more cases than the INT or FLOOR or CEIL function. In this case INT works but there may be other cases where you are not comparing to an integer, and so INT won't work in those cases but ROUND will. ... View more

Maxim 2 ... View more

In case why I am doing this? I need put 'em into one line/macro variable and then attach it to the output dataset. That allows me backtrack what logic indeed is used.   Lots of good ideas above. Most of them involving explaining to us in a lot more detail, or re-inventing the process. You would be wise to follow that advice. As with one of your earlier threads @hellohere there's too much here that makes no sense or are red flags.   Additionally, my advice is that code  generally doesn't belong in datasets, which is another reason to explain more and another reason to consider a different process. And furthermore, I can't see a reason to store code in datasets that doesn't work at all because of imbalanced parentheses). ... View more

@PamG wrote: This is a sample data. What I am trying to get at is that the order in which the variables are entered in PROC SQL when one of them is entered as INTO matters. Is there a way to get COUNT before SUM and assign COUNT to a macro variable in the same PROC SQL code? First, as stated by @Kurt_Bremser we don't know your final goal here. That would be very helpful information.   There are only very rare cases where you need to put statistics like count and sum into a macro variable, and in fact there are better ways to determine these values than SQL. So I think the efficiency of what you are doing could improve, if only we knew what the final goal is. ... View more

The article tells you it comes from someone named Peter Schneider. ... View more

Yes, I'm all set now. I enabled scripts for sas.com and it works. Thanks! ... View more

I see the problem. I have a security plug-in for Firefox called NoScript, it works when I set this to its lowest security level, but not when I set it to the highest security level. ... View more

This is not working in Firefox 143.04 on Windows 11. I click the check box and the web page doesn't change, it still shows the "non-discussion" topics. ... View more