About Andalusia

Pepijn · ‎11-12-2021

I suck at code but if I interpret your question correctly you can sort your data and create a new column using a 'first' function and count the rows within a specific group... sort your data descending (last color on top) count per group with a rank number +1 for every new instance until you encounter the next group (The K-11,K-12 in your data set I think?) And then only filter the rows labelled 1 till 8 and reverse them again if needed. data "have"; SET "output tabel"; by "K-11 colunm name"; RANK +1; if first."K-11 colum name" then RANK = 1; RUN; first sort it the right way, and maybe add other column to your BY group when needed.

ballardw · ‎11-08-2021

"Blue" does not have an automatic numeric value associated with it. So you would have to provide exactly what number you expect for "blue" or "black" or any other color. Also perhaps if you describe how the number for "blue" will be used we can give you some pointers.

Kurt_Bremser · ‎10-20-2021

@Andalusia wrote: Could you show an example? Looks like this: intck('minute',t1.datetime,datetime(),"c") The "c" (for continuous) means that difference / 60 is used, and not the count of minute boundaries between the two values (90 seconds difference can turn out to be 2 minutes if the default is used). For reference, run this: data _null_; time = datetime(); rc = sleep (119,1); diff1 = intck('minute',time,datetime(),"c"); diff2 = intck('minute',time,datetime()); put _all_; run;

Kurt_Bremser · ‎10-13-2021

Create a new dataset, and append it: data have; infile datalines truncover; input type $ model $; datalines; XX-1 Light XX-2 Heavy XX-3 Heavy XX-4 Heavy XX-5 Slim XX-6 Light XX-7 Heavy ; data new; infile datalines truncover; input type $ model $; datalines; XX-8 Light XX-9 Slim ; proc append base=have data=new ; run;

Kurt_Bremser · ‎10-04-2021

The presence of semicolons in the data can be alleviated by using the DATALINES4 statement and 4 consecutive semicolons to end the datalines block. For finding substrings in delimited strings, use the FINDW function: data have; infile datalines truncover; input type $ q_20211 q_20212 q_20213 last_code :$11. second_last_code $; datalines4; XX-1 1 2 3 oo6;001 oo6 XX-2 1 1 2 k04;oo6 008 XX-3 1 3 1 002;001 009 XX-4 3 4 1 oo5;004 004,002 XX-5 4 2 1 001;k04 001 XX-6 4 3 1 008;001 002 XX-7 2 5 3 001;001;004 005 ;;;; data want; set have; length code_checker $5; if findw(last_code,'k04','; ') then code_checker = "RED"; if findw(last_code,'002','; ') then code_checker = "GREEN"; run;

acordes · ‎09-21-2021

For reports that persist in time it's better to save a sashdat copy of the the tables that draw the report. If tables get unloaded and the report opens again it can recover the tables by loading the sashdat files into memory. It is necessary to have the same name and the name caslib.

ChrisHemedinger · ‎09-01-2021

For a logical Workspace Server to appear in EG, ReadMetadata access to its definition must also be granted to the group of users who need to use it. And as @Kurt_Bremser says, those users also need to have account-level permissions to start a process on the server as (usually) it's their host accounts that are used for the running process. Independent of EG, you can also test server connection by using the ITConfig tool that is included in the SAS Integration Technologies client (always installed on the same machine as EG and other SAS tools).

Kurt_Bremser · ‎09-01-2021

@Andalusia wrote: @yabwon and what if in a few months time I get the data of q_20214? I cant be changing the code every time a new quarter comes in.... That is exactly why you use a long dataset layout; new data is then data and not structure.

PaigeMiller · ‎08-23-2021

@Andalusia wrote: Oke and where is the STATUS05? I want STATUS1 and STATUS05 in one view I didn't show code for STATUS05. I expected you would be able to see what changes were made to the STATUS1 code, and make the same changes to the STATUS05 code.

PaigeMiller · ‎05-10-2021

The first step is always determining in your own mind (i.e. not even SAS code at this point) valid rules that will create these colors. I have added in red modifications to what you typed to produce valid rules counter <= (avg + std * 1) then 'green' counter > (avg + std * 1) and <= (avg +std * 2) then 'yellow' counter > (avg + std * 2) and <= (avg + std * 3) then 'orange' counter > (avg + std * 3) then 'red' From there, SAS code is relatively easy to produce. data want; set have; length color $ 6; if counter <= (avg + std * 1) then color= 'green'; else if counter > (avg + std * 1) and counter <= (avg +std * 2) then color='yellow'; else if counter > (avg + std * 2) and counter <= (avg + std * 3) then color= 'orange'; else if counter > (avg + std * 3) then color= 'red'; run; There are some minor simplications you can make to the above, such as changing the next to last line to else color='red';

Kurt_Bremser · ‎05-04-2021

Add a final MERGE step to your code: proc summary data=extract; by type phase; var counter; output out=want1 (drop=_type_ _freq_) mean()=mean std()=std; run; data want2; merge have want1 (in=e) ; by type phase; if e; run;

Andalusia · ‎04-28-2021

@PaigeMiller @Kurt_Bremser Thank you both. That was indeed the last step I missed!

Ksharp · ‎04-22-2021

Post some sample data. So I can replicate your question.

ballardw · ‎04-21-2021

@Andalusia wrote: @ballardw in my real dataset I have much more points. Just a question. What if I want to calculate the AVG and STD of based on the last 4 quarters instead of 2? The Lag approach I suggested for two cases can be extended back to 100 records. However looking back that way adds in potential issues with looking back into a previous phase or type. You are adding in some complications that raise questions of 1) will there always going to be at least 4 quarters? 2) if there are fewer than 4 quarters but you want 4 what to do, 3) Are your "4 quarters" consecutive, if not how to treat gaps? Also is "last" actually best description? Quarters usually relate to calendar dates and do you mean "after a given date" instead? (This is much easier to do once actual dates are supplied in data). Make example data that demonstrates more of the actual data and replicating and problem cases such a fewer "quarters", missing quarters if they ever occur in the data (you have looked haven't you). Then show the desired values for that example data. Your "example" data might imply that 4 quarters means a calendar year. In which case construct a calendar year variable and use proc means or summary with type, year and phase as class variables and request the statistics for the variable as a VAR variable. Example: data need; set have; year = int(quarter/100); run; proc summary data=need nway; class type year phase; var counter; output out=work.summary mean= std= /autoname; run; If your last 4 quarters crosses a calendar year boundary then you need to show example data and possibly provide a cleaner description. BTW, consider making your quarter variable a date value because then SAS provides a number of tools that are easier to work with than pseudo-date values that need parsing for every step.

Andalusia · ‎04-21-2021

I ran your snippet and merged it with the transposed table, but I miss 2 essential columns that I had in my HAVE dataset. I miss diff_phase and ceation_date.

Online Status	Offline
Date Last Visited	‎11-08-2021 07:20 AM

Re: How to convert a character to a numeric value?

How to convert a character to a numeric value?

Re: How to get the last 8 values in a group?

How to get the last 8 values in a group?

Re: How to calculate the difference between two datetimes in minutes?

How to calculate the difference between two datetimes in minutes?

How to add new data to an existing dataset in EG?

Create a new table based on an older one

Re: How to get the last value?

Re: How to get the last value?

Re: How to get the last value?

Re: How to properly use IF/ELSE statements in SAS

Re: How to properly use IF/ELSE statements in SAS

Re: How to use IF/CASE statements in SAS EG?

Re: How to merge two calculated datasets in SAS EG?

Re: How to get the last 8 values in a group?

Re: How to convert a character to a numeric value?

Re: How to calculate the difference between two datetimes in minutes?

Re: How to add new data to an existing dataset in EG?

Re: Create a new table based on an older one

Re: Cant change a data source in SAS Visual Analytics

Re: How to add a new server in SAS EG?

Re: How to get the last value?

Re: How to properly use IF/ELSE statements in SAS

Re: How to use IF/CASE statements in SAS EG?

Re: How to merge two calculated datasets in SAS EG?

Re: How to calculate variables inside a group?

Re: How to make calculations inside a group?

Re: How to calculate the average and standard deviation of the last 2 ...

Re: How to get the latest and second latest value in a group?