Agreeing with @Tom that numbers should not characters.   Once you perform the conversion to have numeric values, PROC SUMMARY or PROC MEANS produces the dataset or output you want. ... View more

SAS has never (as far as I know) used the normal distribution when there are more than 30 data points. It uses the t-distribution even when there are 30,000 data points (or more) assuming of course that the t-distribution is the correct distribution for the problem.   The idea of using the normal distribution when there are more than 30 data points is an approximation, and was useful back in the days before computers and you had to use a look-up table. But it isn't really needed with modern computers. ... View more

Are you using PROC VARMAX? It would appear that the parameter estimates follow a t-distribution. According to the documentation, the parameter estimates converge to a multivariate normal distribution as T approaches infinity (T apparently is the number of data points), which means before they converge they are likely a t-distribution. ... View more

As I said in another one your suggestions, I don't think this belongs in Suggestions. When some numbers aren't displayed as you want, you could simply apply the right format/function to the data before displaying it. SAS already gives you all the tools to display dates, datetimes and times in whatever way you want.   So if you have a variable named DATETIME which contains year, date, month, hours, minutes, seconds and milliseconds, you could create a new variable which contains the time part and format it as you wish.   In a SAS data step   hour_min = timepart(datetime); format hour_min hhmm.; ... View more

I'm not sure this belongs in the Suggestions section. It seems easy for you to change. (And by the way, you don't say how your data set is created, but it may be that whoever created the data set included the commas for some reason. There is really nothing in SAS that forces these numbers to have commas.)   If these are numeric, you can simply change the format to remove the commas. If these are character, you can use the COMPRESS function to get rid of the commas. ... View more

@Tecla1 wrote: Good Morning.... Tnks fo your replay... I receive this error: ERROR: Insufficient authorization to access /Totale_campioni_polizze_addebiti__&des..XLSX.   This is why you ALWAYS need to include the log when you have code the doesn't work. There's no way we could have guessed from your code and extremely brief description that this is the problem. And we want not just the error message, but the code in the log that precedes the error message, you need to show us at least one PROC or DATA step in the log before the error message.   I think @Quentin has the right answer. ... View more

Posting code that doesn't work, without a log, is usually not productive. ALWAYS show us the log. And we need to see not just the error messages; we need to see the preceding code as well, covering one whole PROC or DATA step before the error message.   Also, the description of the problem is way too brief: "I'm not able in any way to put more sheet in the same file."  ... View more

SAS offers many tutorial videos about SAS Viya. There's even one called "SAS Viya Quick Start".   https://video.sas.com/category/videos/sas-viya-quick-start   ... View more

Is there an option set at invocation or in an AUTOEXEC file or at the top of the program to use an external (to your program) macro library? For example:  options sasautos=(sasautos 'G:\xxxx\xxxxxxxxxx\xxxxxxx\Users\MillerP\Macros');   This allows users to use any macro in my Macros folder, without including the macro in the program itself. Of course, in your case the command would use a different folder than I showed above. If that macro library has moved or been deleted, you would get the error you showed. If there is no such option in your invocation or autoexec or at the top of the program, you would get the error you showed. Maybe this option should be added to your code, if you can find where the macro library is located.   The log you show from an earlier run produces PROC REPORT code, that must be coming from somewhere. Most likely — almost definitely — it is coming from the now unavailable macro named %create_state_tab. You say it was never interpreted as a macro, but the old log that you provided makes it clear it was a macro.   It would help diagnose and debug macros if you added the option MPRINT to your autoexec or at the top of the calling program. I have it always turned on, as it doesn't hurt if you don't need it, but it is extremely helpful when creating and running macros. In your autoexec, use    options mprint; ... View more

Agreeing 110% with @Quentin . Reports are not meant for additional SAS data analysis, even if you choose to store them as a SAS data set. We gave you advice in your previous thread, you ignored it, and so now you have additional problems that could have been avoided had you followed our advice. Unfortunately, we see that from many SAS users. It would be better if you kept the data in proper SAS data set form so you could continue to analyze it, and also created the report (perhaps saved as a second SAS data set, although this is not required or recommended).   Your problem description appears to me to be incomplete. Are we only changing the one row with Class1=. directly under Class1='Blood_disease'? We only change this one row to have the Value in the row above? Could there be more than one row under Blood_disease? Do we ever need to do this with Lung_disease? Are there ever any other diseases mentioned in Class1? If so, what do we do with these? Does your real data have only one patient, or are the multiple patients in your data set that we will need to do this? If there is more than one patient, how would that data set look? ... View more

Your data does not match the problem description where you said   Complete_response = 5 pts Overall = 6 pts. Rate = 5/6 pts = 0.83. Note that the numerator is included in the denominator.   Sometimes you have to modify the data set to match the description, or vice versa.   Your analysis uses the WEIGHT statement and so it is calculating results based on 11 "units" (5+6=11) and a proportion of 5/11, but that doesn't sound like what you want. ... View more

Can you explain why a dataset has to have a blank line?   It makes perfect sense to have an output from PROC REPORT or PROC PRINT with a blank line. I can't think of a reason why a dataset needs a blank line. ... View more

@SASdevAnneMarie wrote: Thank you, Tom. I'm wondering if I can do this with "prxmatch" : %macro mymacro(var); if (prxmatch("/\b0[,\.]&VAR.\s*%/",lb_lg)>0 then .... %mend; %mymacro(var=50); Your are wondering if you can do this with prxmatch. The answer is for you to try it.  it is impossible for me to advise you further since I don't know what you are trying to do. Would you please explain what you are trying to do? Please explain in words without reference to PRXMATCH or any other code.   ... View more

@SASdevAnneMarie wrote: Thank you, Paige Miller, but I didn't fully understand your idea. The whole idea is that numbers are easier to handle as numbers than if you leave them as text. If there was something I could advise less experienced people to do, it would be to ALWAYS convert text strings which are really numbers to actual numbers.  ... View more

I think far easier is to convert the text which you describe as 0.50% to 2.70% to actual numbers and then checking becomes much much much much easier. And then likely no macro is needed. ... View more