About RoddyJ

ballardw · ‎10-04-2022

@RoddyJ wrote: Thanks for the answer, this hasn't worked I'm afraid. I've tried changing up the order of options and ODS PDF and changing the paper size but sadly no joy. Time to share sample data that will recreate the problem and complete code, such as the original ods pdf statement. if you are attempting to change from portrait to landscape in the middle of the document that may not work.

FreelanceReinh · ‎06-22-2021

Hi @RoddyJ, I would check the default length of format $teamf.: proc format fmtlib; select $teamf; run; Output starts like this: ---------------------------------------------------------------------------- | FORMAT NAME: $TEAMF LENGTH: 20 NUMBER OF VALUES: 3 | | MIN LENGTH: 1 MAX LENGTH: 40 DEFAULT LENGTH: 12 FUZZ: 0 | If the default length is smaller than the length (as shown above), truncation will occur for longer values (unless you specify a greater length in the PUT statement as suggested by Tom; this length is limited by the maximum length, 40 in the example). The default length might have been set inappropriately by a variable named default in dataset neonate_formats which has too small a value in the first observation for format $teamf..

RoddyJ · ‎05-28-2021

Thanks Tom, this is a very good solution!

JeffMeyers · ‎05-20-2021

Two different things I would look at here. What is hopefully the easier solution is if your x-axis is being turned into a discrete axis (which it looks like). This has the behavior with VALUES to remove the values not in the list. From your X-axis statement it looks like you'd rather have a linear axis potentially. If you add TYPE=LINEAR to your x-axis statement it might just make the fix with your current VALUES statement: proc sgplot data=sashelp.class; vbar AGE / response=height; xaxis type=linear values=(11 to 15 by 2); run; Otherwise if you need to keep it as a discrete axis, then it gets more tedious. There is another option called VALUESDISPLAY where you can specify the printed value for each tick on the axis so long as they are listed in the VALUES statement. If you have a lot of bars you would probably want to use macro language to do this: proc sgplot data=sashelp.cars; vbar type / response=msrp group=origin; xaxis values=('Hybrid' 'SUV' 'Sedan' 'Truck' 'Wagon') valuesdisplay=('Hybrid' '' 'Sedan' '' 'Wagon'); run;

Kurt_Bremser · ‎11-30-2020

Cartesian join in SQL: proc sql; create table want as select ds1.cat1, ds2.cat2 from ds1, ds2 ; quit;

RoddyJ · ‎04-06-2020

Thanks for this reply, really well explained!

Tom · ‎02-21-2020

Show an example of what is wrong. It is not clear what variable you are talking about. You don't show a variable named TIME. Your code generates dataset TIME_DIFFS, MEDIAN_DIFFS and DIFF_OUTPUT. Which of these are you exporting? Your code mentions variables named T1, T2, DIFF, BY_VAR, and DIFF_TEXT. Which of these are showing properly? Which are not?

RoddyJ · ‎01-07-2020

Yes I made a mistake in my original post, now amended. Thank you for pointing that out and for your solution!

DanH_sas · ‎12-06-2019

The DATALABELPOS option is basically a convenience option I added to do what I showed you with an AXISTABLE. With that convenience, I have to make some decisions internally. One of those involves labeling. If there is no GROUP, there is no point in showing a label, so no label appears. But, for a GROUP situation, I normally need to show a label so that you can identify each set of values per group value. For CLUSTER situations, however, that label is not as useful. Just know that XAXISTABLE/YAXISTABLE statements will give you the most control when creating "tabular" labeling.

gamotte · ‎11-28-2019

Hello, count function gives you the number of observations for the variable in the resulting dataset, no matter what particular values the variable takes. data have; input dummy; cards; 0 1 1 0 0 0 1 0 ; quit; proc sql; SELECT count(dummy) FROM have; quit; Your condition (pta = "Yes") creates an implicit dummy variable as the one in the example above. When pta <> "Yes" the observation will be zero but will be counted anyway. => You have to use the sum function as @PeterClemmensen suggests.

Tom · ‎11-14-2019

SAS only has two data types. Fixed length character strings and floating point numbers. Note that a SAS format is just instructions for how to convert the values to text when displaying them and does not define a new type of variable. If sounds like you have a numeric variable that has a format attached to it. When you make the new variable just the values are copied. You just need to attach a format to the new variable so that it will look to you like a date. Use the same format that was attached to the original variable if you want the values to display in the same way. data times; x_time = coalesce(date1,date2); format x_time datetime20.; run; You might want to change your variable names. From the format you have attached it appears that all three of those variables contain numbers that represent DATETIME values. Yet your names imply that two of them have DATE values and the third as a TIME value.

PaigeMiller · ‎09-19-2019

So @Kurt_Bremser has pointed out the error, but it might be helpful to follow certain advice before you try to turn code into code with macro variables and/or code with macros. The advice is to get the code to work on one or two situations without macros and without macro variables. If your code doesn't work without macros and without macro variables, then it will not work if you are using macros and macro variables. If it does work without macros and without macro variables, then you have a good chance it will work with macros and with macro variables.

RoddyJ · ‎09-09-2019

Thanks so much for your help on this, really appreciate it.

Online Status	Offline
Date Last Visited	‎10-05-2022 02:00 PM

Re: SGPLOT output being cut off in ODS PDF

SGPLOT output being cut off in ODS PDF

Variable created with PUT format not correct length

Re: Converting array of numeric columns to character columns

Re: Converting array of numeric columns to character columns

Re: Converting array of numeric columns to character columns

Re: Converting array of numeric columns to character columns

Re: Converting array of numeric columns to character columns

Converting array of numeric columns to character columns

Remove bar chart tick labels but not ticks

Re: Variable created with PUT format not correct length

Re: Variable created with PUT format not correct length

Re: Remove bar chart tick labels but not ticks

Re: Converting array of numeric columns to character columns

Re: Remove bar chart tick labels but not ticks

Re: SGPLOT output being cut off in ODS PDF

Re: Variable created with PUT format not correct length

Re: Converting array of numeric columns to character columns

Re: Remove bar chart tick labels but not ticks

Re: Merge every observation in one dataset to each observation in anot...

Re: Remove duplicate observations based on multiple criteria

Re: Formatted variables incorrectly outputting to excel

Re: Can this be done using proc transpose?

Re: Issue with hbar datalabel

Re: SQL Count not counting

Re: How do I Convert Numeric type to Date type?

Re: Variable defined in %let won't work in WHERE statement

Re: How to Label Plot Axis with Macro Variable Name

SAS Analytics Explorers