A journey of a thousand miles begins with the first step.
In the workspace there is a semantic model called “Province Population”. The report for this semantic model has been deleted.
When I try to delete the model, I got the following error message.
I need to unpublished app in the workspace on the PowerBI service.
Then I can delete the semantic model from the workspace.
Once the app is unpublished the “Create app” menu also shows.
I got the following error when opening the SAS Enterprise Guide 7.1 (64 bit).
‘.’, hexadecimal value 0x00, is an invalid character. Line 1, position 1.
I found the following related post on the SAS community website, https://communities.sas.com/t5/SAS-Enterprise-Guide/Error-when-launching-SAS-EG-7-1/td-p/403943 with reference to the SAS knowledge base https://sas.service-now.com/csm?id=kb_article_view&sysparm_article=KB0036279. Here is the solution proposed in the knowledge base.
Workaround
To resolve the issue, follow these steps:
The solution worked as of November 2024.
Value of macro variable id has been assigned in the local session. The macro variable is required for the remote session. Use %syslput to assign the value to the local macro variable to the sever session (remote).
Reference: https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/connref/n1bc4zi0lp02ivn1oxn9gc1bsaha.htm
%syslput: Creates a single macro variable in the server session or copies a specified group of macro variables to the server session.
%put &id;
%syslput id=&id;
rsubmit;
%put &id;Code:
ods listing close;
ods html file="X:\Work\UG_AWARD\Validate\2223awards_oldcode.xls" style=journal;
proc tabulate data=hist.awards2223_sas missing f=dollar11. noseps;
class prog_fac level source level2 category awardname;
var awardamt;
tables level,source*level2*category all,awardamt=' ' / rts=40 indent=3 misstext='$0';
tables level,source*level2*category all,(prog_fac='Home Faculty' all)*awardamt=' ' / rts=40 indent=3 misstext='$0';
*tables level,source*level2*category*(awardname all) all,(prog_fac='Home Faculty' all)*awardamt=' ' / rts=40 indent=3 misstext='$0';
keylabel sum=' ';
run;Error:
ERROR: A component of C:\Users\username\AppData\Local\Temp\SAS Temporary
Files\computer_name\X:\Work\UG_AWARD\Validate\2223awards_oldcode.xls is not a directory.
ERROR: No body file. HTML output will not be created.
Reference:
SAS Support Usage Note 61280: The ODS HTML statement generates errors in the SASĀ® 9.4 TS1M5 windowing environment when the FILE= option includes a fully qualified path
Fix:
Cause: The problem occurs because the “Use WORK folder” option is set as the default in the SAS registry and preferences. This causes the WORK folder to be prefixed to the path that is specified with the FILE= option. Separating the File and Path options can also fix the problem
A. Academic Faculty
Statistics on full-time academic teaching staff at Canadian universities: Interactive tool
Student Pathway
Persistence and graduation of postsecondary students aged 15 to 19 years in Canada: Interactive tool
Labour Market Outcome
Labour market outcomes for college and university graduates: Interactive tool
Education Indicator
Education Indicators in Canada: Report of the Pan-Canadian Education Indicators Program
In previous 3 Proc Tabulate related posts, I have focused on table coloring, percentages, and setting order for the class variable .
In the table coloring post, I have provided example of stacking statistics for multiple row variables one after the other. Sometimes, we the variables has a nested relationship, we can use * to identify the hierarch of the variable, eg Faculty*program so the statistics by program are ordered by Faculty and then program.
proc tabulate data =table (where=(exclude = "NO" )) missing; class visa fullpart gender program faculty year; var heads fftes; table faculty="Faculty" all visa="Visa Status" all fullpart="F/P" all gender="Gender" all faculty="Faculty"*program="Program" all, (year="year" all)*(heads="Heads"*SUM="" heads="% Total"*pctsum fftes="FFTEs"*SUM="" fftes="% Total"*pctsum < faculty all visa all fullpart all gender all faculty*program > ="" ); run;
The above code basically stacking the sum and percentage of heads and FFTEs by year (column) for the row variables in the order of Faculty, Visa, fullpart (full-time or part-time), gender, and Program (nested under Faculty).
Please pay attention to the “missing” option in the Proc tabulate. If there is any missing value in the class variables, the whole observation is removed from the Proc tabulate result, even for the class variable that doesn’t have any missing values, because Proc tabulate force the total statistics for all the class variables to be the same. It is crucial to include the “missing” option, so “” has a value to be included in the Proc Tabulate results.
Here is a good reference document for Proc Tabulate.
Reference: https://www.iuj.ac.jp/faculty/kucc625/sas/tabulate.html
proc tabulate data= sashelp.cars;
class origin type make drivetrain;
var msrp;
table (origin="" all) * (make="" all), (type="" all )*(DriveTrain="" )*(MSRP=""*MEAN N);
run; Output:
The table produced will show a hierarchical summary:
Delete $year format from sasfmt.commonfmt catelog.
proc catalog catalog =sasfmt.commonfmt ; delete year.formatc / entrytype=format; quit;
Assume there is a $year format in the work library. Copy the $year format to the sasfmt.commonfmt.
The best practice is developing a local permanent format catalog for common usage, in this case the sasfmt.commonfmt. If I need to create new format for the SAS work and the format is reusable, then I will copy the format from work library to the sasfmt.commonfmt catalog so I don’t have to recreate the format in the future.
Proc catalog catalog=work.formats; copy out=SASFMT.COMMONFMT; select year.formatc; quit;
Check the $Year format in the sasfmt.commonfmt catalog.
proc format fmtlib lib=sasfmt.commonfmt ; select $year; run;
Normally I need to compile a unique list of text strings from SAS table and run report for each item in the list through a %do loop. See details in the previous post. In this case, I need to convert the list of strings back to SAS table using space as delimiter.
Solution
%let lst = apple orange grape carrot banana watermelon peach; %macro createtbl; data lst; length fruit $15.; %let i = 1; %do %while (%scan(&lst., &i., ' ') ne ); fruit = "%scan(&lst., &i. , ' ')"; output; %let i = %eval(&i. +1); %end; run; %mend createtbl; %createtbl;
Notes
Wrong Code Demo and Error Diagnosis
data lst; length fruit $15.; do i =1 to 7; fruit = "%scan(&lst., i , ' ')"; output; end; run;
Error Message
Diagnosis
Other Methods that don’t work
/* NOT USEFUL*/
data lst;
attrib
fruit length=$15 format=$15. label="fruit";
stop;
run;
Since our office got us both the Tableau and Power BI license, it is up to us to incorporate these data visualization tools in our analytical work. It takes a while to figure out how to perform particular tasks in both Tableau and Power BI, though the learning curve for Tableau is steeper, but once you figure out a way to accomplish what you suppose to do with the software, the efficiency and the output are quite amazing.