Currently at my job, I use PC-SAS 9.4, 32-bit and Microsoft Excel 2010, 32-bit. The company is considering moving to Microsoft 365, which is a 64-bit version. This could be problematic, as in my job, importing and exporting between Excel and SAS is done a lot, with bringing in information from multiple tabs in excel files to manipulate data from another source (database), to then export to multiple workbooks.   I am, however, extremely unfamiliar with the whole 32-bit and 64-bit concept and am looking for information on compatibility issues that could occur between 32-bit SAS and 64-bit Microsoft Office/Excel. In particular, how would it affect using the PROC IMPORT and PROC EXPORT statements between the two, given the excel files are .xlsx and .xlsm (workbook and macro-enabled) formats. Any information in appreciated, thank you. ... View more

Hello all,   I'll try to explain this as best as possible. I have a SAS Dataset with just one column, that has a list of about 3 to 4 million unique IDs. On the teradata server, there is a table whose data I want to pull for all the IDs in the SAS dataset. The issue I am encountering is that when I use PROC SQL, it does not recognize or seem to know about the SAS Dataset, thus I can't do a left (or right) join on the pull.   I also can not do a pull on just the table from the teradata server as it is extremely large and the only condition I have to limit the pull are these IDs. Here is an edited sample of the program I am running in what I am trying to do, with the SAS Dataset named SAS_TABLE and the teradata table named TERA_TABLE:   PROC SQL; CONNECT TO ODBC(DATAsrc='SERVER NAME'); CREATE TABLE PULL1 AS SELECT * FROM CONNECTION TO ODBC ( SELECT * FROM SERVER.TERA_TABLE B RIGHT JOIN SAS_TABLE A ON A.ID = B.ID ); QUIT; I've tried it this way and also a left join version, but each time, I get an error stating the SAS_TABLE does not exist:   ERROR: CLI prepare error: [Teradata][ODBC Teradata Driver][Teradata Database] Object 'SAS_TABLE' does not exist   I am assuming what the program is trying to do is find "SAS_TABLE" in the actual database, which is what is causing this. Any help on this would really be appreciated, thanks! ... View more

Hello,   I am pulling data from a database using PROC SQL and there is a WHERE condition I am unsure on how to input. The pull is one table with several columns (variables) and I want to exclude rows when two of the columns have a specific value each.   As an example, call the columns Alpha and Beta. If in one row, Alpha has the value C and in that same row, Beta has the value X, I do not want that row in the pull. But if Alpha has the value C and Beta has any other value but X, or if Beta has the value X and Alpha has any other value but C, I do want those rows. Any help on this is appreciated, thanks! ... View more

I have a very large table with ID numbers and dates associated with them. As as example of a table with two columns (I will use names instead of ID numbers for simplicity):   John    1/1/2010 John    1/1/2011 John    1/1/2012 John    1/1/2013 Sam     6/1/2012 Sam     4/1/2013 Sam     4/1/2014 Victor   7/1/2011 Zack    4/1/2011 Zack    4/1/2012 Zack    10/1/2012 Zack    10/1/2013   From this table, I want to make two tables - One that has the name and the earliest date associated to that name (member ID) and one that has the name and the last date associated to that name. Thus the first table would come out as:   John    1/1/2010 Sam     6/1/2012 Victor   7/1/2011 Zack    4/1/2011   And the second table (with the last date) would come out as:   John    1/1/2013 Sam     4/1/2014 Victor   7/1/2011 Zack    10/1/2013   I've included the example of "Victor" as there are records with just one date, thus the same date would have to count as both the first and last. Thank you to any that see this and can help! ... View more

I am pulling data from a Teradata using PROC SQL on a PC-SAS (SAS 9.4 TS Level 1M3) platform. One of the issues I am having is with the date format. In the Teradata, the date is formatted as '2016-01-15' and when I pull it, SAS converts it to a 15JAN2016 format (DATE9. format). I need to convert the date to a YYYYMMDD style (20160115), which I believe would be a character variable (honestly don't know), before exporting the data to a text file.   This is the PROC SQL (I've edited the table and column names for simplicity):   %LET DATE1_ST_DT = '2018-01-01'; %LET DATE1_END_DT = '2018-01-31'; %LET DATE2_ST_DT = '2016-01-01'; %LET DATE2_END_DT = '2018-02-01'; PROC SQL; CONNECT TO ODBC(DATAsrc='RANDOMDB'); CREATE TABLE EXAMPLE AS SELECT * FROM CONNECTION TO ODBC ( SELECT A.DATE1, B.DATE2, C.DATE3, C.COSTS FROM ASYSTEM.TABLE1 A INNER JOIN ASYSTEM.TABLE2 B ON A.ID_MATCH1 = B.ID_MATCH1 AND A.ID_MATCH2 = B.ID_MATCH2 INNER JOIN ASYSTEM.TABLE3 C ON B.ID_MATCH3 = C.ID_MATCH3 AND B.ID_MATCH4 = C.ID_MATCH4 AND B.ID_MATCH5 = C.ID_MATCH5 WHERE A.DATE1 BETWEEN &DATE1_ST_DT AND &DATE1_END_DT AND B.DATE2 BETWEEN &DATE2_ST_DT AND &DATE2_END_DT ); QUIT;   Ideally I would like to be able to change the format in the same PROC SQL sequence, so that the resulting date variables are in the YYYYMMDD format, thanks! ... View more