I think this is a fair critique.  SAS documentation is excellent.  But I do feel like with SAS 9 clients (EG, DI Studio, Stored Processes), SAS has added in a lot of macros to provide functionality (%_eg_whereparam, %stpbegin, etc) without always documenting them in the same way they would document a new PROC or STATEMENT.  Some of them you can't even find the source code for. My guess is many of these were designed to be used 'behind the scenes' by SAS developers, or used by SAS clients written by SAS developers.  It feels like they weren't really written with the expectation that us folks outside of SAS would start using them in our own code.  But, that was probably short-sighted. If you search lexjansen.com for %_eg_whereparam, you will find papers written by the SAS community that explain it a bit more, including some by SAS employees, e.g. https://support.sas.com/resources/papers/proceedings11/309-2011.pdf     ... View more

thanks a ton ... View more

It's in Tasks - Data - Copy Files @mrdlau wrote: I can move the file to the citrix location.............   how do I use the built in file transfer task?   ... View more

Easiest method is to not use characters as your X axis. Instead use numeric values and then use a format to have them displayed for presentation.    Because its a categorical variable you can also use the VALUES = option on the XAXIS statement to have the order you want.   http://documentation.sas.com/?docsetId=grstatproc&docsetTarget=p07m2vpyq75fgan14m6g5pphnwlr.htm&docsetVersion=9.4&locale=en#n09f4ln7jq8gkan1w2nhlkzjcrqz   proc sgplot data=sashelp.cars; xaxis values=("GMC" "Honda" "Hyundai") offsetmin=0.2 offsetmax=0.2; scatter x=make y=mpg_city; run; PS. please take some time to go through your historical posts and mark them as answered/solved with the correct response tagged.   @mrdlau wrote: I'm generating a simple vbar chart with the Y axis being counts, and x axis being ranges that I created as text (i.e "0 to 5%", "5.01 to 10%" etc).   When I plot the chart,  my X axis is out of order.  it'll go "0 to 5%", "10 to 15%", "5 to 10%".  How do I set the order in this scenario where the x axis is text? Proc sgplot data=gridwork.data; vbar dist/ response=counts; xaxis label=distribution; yaxis label='counts; run; quit;   ... View more

&path has to refer to the path and file you're importing. The format depends on your OS. ... View more

This might work for you:   data a; do y = 1990, 2000, 2010, 2020, 2025, 2040; output; end; run; data b; do y = 1991, 2002, 2015, 2030; output; end; run; proc sql; select a.y, b.y as closest_y from a left join b on b.y >= a.y group by a.y having min(b.y) = b.y; quit; y closest_y ------------------- 1990 1991 2000 2002 2010 2015 2020 2030 2025 2030 2040 . ... View more

@mrdlau wrote: I have a variable z=10, but in a character format. To multiply it by 2, i convert the z to a numeric during my multiplication: 2*input(z, 8.2). However, my result is .2, not 20. Why is this the case? It is onlu 20 when I use input(z, 8.) You might try a simple   x = 2*z; and examine the result and the log. SAS will attempt to convert character  values to numeric to do computations like this in a data step. The conversion is usually pretty good. Though if you want to do calculations the question is always why was the variable created as character in the first place. ... View more

If you want to reference a derived value in the same SQL statement use the CALCULATED keyword. proc sql; select sum(fee1,fee2,fee3) as Fee123 , sum(fee4,fee5,fee6) as Fee456 , sum(calculated Fee123,calculated Fee456) as TotalSUM from .... ; quit; You do realize you are using the SAS function SUM(,) and not the SQL aggregate function SUM() in that query? So it is calculating separate sums for each observations. It is similar to if you coded use + operator, but with missing values treated as zero. fee1+fee2+fee3 as Fee123 , fee4+fee5+fee6 as Fee456 ... View more

Firstly , set the cell of excel be TEXT attribution, and then Past the number into that cell . ... View more

If you require "No" for display purposes then a format would do a good job:   proc format; value no 0 = "No"; run; proc sql; select t1.variable as ColumnName format = no. FROM T1; quit; ... View more