About radhikaa4

radhikaa4 · ‎11-16-2020

Hello. I am trying to merge two datasets. What I would like to do is full join per subject ID even though date_time from dataset 1 doesn't exisit in dataset 2; First dataset: SubjectID date_time value1 123 1/1/2010 9:53 52.1 123 1/2/2010 10:31 21.2 123 1/3/2010 9:33 32.3 456 2/1/2010 9:53 42.2 456 2/2/2010 10:31 123.4 456 2/3/2010 9:33 47.2 456 2/4/2010 14:23 44.4 second dataset: SubjectID date_time value1 value2 123 1/1/2010 9:53 52.1 10.2 123 1/2/2010 10:31 21.2 42.1 456 2/1/2010 9:53 42.2 44.8 456 2/4/2010 14:23 44.4 32.2 want: SubjectID date_time value1 value2 123 1/1/2010 9:53 52.1 10.2 123 1/2/2010 10:31 21.2 42.1 123 1/3/2010 9:33 32.3 456 2/1/2010 9:53 42.2 44.8 456 2/2/2010 10:31 123.4 456 2/3/2010 9:33 47.2 456 2/4/2010 14:23 44.4 32.2 Here's my code so far: proc sql; select t1.subjectid, t1.date_time, t1.value1, t2.value2 from dataset1 t1 left outer join dataset2 t2 where t2.value2 is null any help would be greatly appreciated

radhikaa4 · ‎11-13-2020

Thank you! Just a quick question - I tried the second block of the code, but I am getting error at " b) ) c" . it says syntax error but I tried to remove ) and add it else where but still doesn't work

radhikaa4 · ‎11-12-2020

Hello. I have two datasets- what I would like to do is that combine two dataset by group and month_year but also include the ones that do not overlap with dates: Dataset A month_year expected_rate 1/1/2017 0 2/1/2017 5 3/1/2017 10 4/1/2017 15 5/1/2017 35 6/1/2017 45 7/1/2017 51 8/1/2017 53 9/1/2017 59 10/1/2017 65 11/1/2017 95 12/1/2017 100 Dataset B group actual_rate rmd_my A 2 2/1/2017 A 7 3/1/2017 A 51 10/1/2017 A 67 11/1/2017 B 0 1/1/2017 B 4 2/1/2017 B 10 4/1/2017 B 25 5/1/2017 B 65 12/1/2017 WANT: group month_year expected_rate actual_rate A 1/1/2017 0 A 2/1/2017 5 2 A 3/1/2017 10 7 A 4/1/2017 15 A 5/1/2017 35 A 6/1/2017 45 A 7/1/2017 51 A 8/1/2017 53 A 9/1/2017 59 A 10/1/2017 65 51 A 11/1/2017 95 67 A 12/1/2017 100 B 1/1/2017 0 0 B 2/1/2017 5 4 B 3/1/2017 10 B 4/1/2017 15 10 B 5/1/2017 35 25 B 6/1/2017 45 B 7/1/2017 51 B 8/1/2017 53 B 9/1/2017 59 B 10/1/2017 65 B 11/1/2017 95 B 12/1/2017 100 65 I tried the following proc sql command: Proc sql; SELECT * FROM dataset_a FULL JOIN dataset_b ON dataset_b.month_year= dataset_a.month_year I also tried full outer join, right join. None of them seem to work

radhikaa4 · ‎06-03-2020

Hi I have the following dataset What I would like to do is to capture min two date time values, if missing use whichever one is available ID first_stop_date_time second_stop_date_time 1234 01FEB20:21:56:00 01FEB2020:13:15:00 4567 02FEB20:15:09:00 03FEB2020:05:12:00 8910 21FEB20:00:00:00 New dataset ID first_stop_date_time second_stop_date_time min_stop_date_time 1234 01FEB20:21:56:00 01FEB2020:13:15:00 01FEB2020:13:15:00 4567 02FEB20:15:09:00 03FEB2020:05:12:00 02FEB20:15:09:00 8910 21FEB20:00:00:00 21FEB20:00:00:00 The only way I can think of is transpose and combine into one date field and then use min. Was wondering if there is another easier way

radhikaa4 · ‎05-18-2020

Thank you. Yes that works but I know it typically in histogram, built in frequency % is generated based on the x_axis. However, I want to choose my own y_axis by x_axis frequency broken down by yes/no colors.

radhikaa4 · ‎05-18-2020

Hi My current dataset looks like this: ID start_date_time stop_date_time discharge 1234 4/4/2020 12:42 4/9/2020 12:11 5 4567 8/1/2019 19:12 . 8910 8/2/2019 18:12 8/11/2019 18:12 9 WHat I would like to do if discharge < 7 then new_stop_date_time = stop_date_time else if discharge >= 7 then new_stop_date_time = stop_date_time else if discharge = . then new_stop_date_time = start_date_time + 7 days I tried the following data o.time_to_first_event2; set o.time_to_first_event; if discharge < 7 then new_stop_date_time= stop_date_time; else if discharge >= 7 then new_stop_date_time= stop_date_time; else discharge=. then new_stop_date_time= intnx('dtday', start_date_time, 7); format new_stop_date_time DATETIME16.; run; Not working! discharge is in BEST12. format

radhikaa4 · ‎05-15-2020

yes, i need frequency by y_axis and then # of patients who were in that bin. i added 'freq y_axis" not working proc univariate data=work.data; class label_category; var x_axis; freq y_axis; histogram x_axis/ normal overlay VAXISLABEL = "Volume trend" ; label label_category= 'Label Category, 1 = yes, 0 = no"; label x_axis= 'Volume trend'; run;

radhikaa4 · ‎05-15-2020

I have one datetime variable and I would like to add 7 days to the existing date/time patientID raw_start_date_time new_start_date-time 1 12/4/2019 15:21 PM 12/5/2019 15:21 PM 2 12/5/19 12:42 PM 12/12/19 12:42 PM I did the following and didn't work new_start_date = intck ('dtday', raw_start_date_time, 7) I also tried intck('day'), intnx ('day'), and raw_start_date_time * 7 * 24 * 60 * 60. No luck

radhikaa4 · ‎05-13-2020

Hello i have the following dataset. What i would like to do is create two overlay histograms by label_category with x_axis being x axis and use y_axis as y axis. patientID y_axis x_axis label_category 1 51 5.2 0 2 30 5.4 1 3 98 1.2 1 4 93 1.7 0 5 17 14.1 0 Here is what I have so far but i am not sure where to add y_axis proc univariate data=work.data; class label_category; var x_axis; histogram x_axis/ normal overlay VAXISLABEL = "Volume trend" ; label label_category= 'Label Category, 1 = yes, 0 = no"; label x_axis= 'Volume trend'; run;

radhikaa4 · ‎03-03-2020

Hi @yabwon . Thanks. How do I include ID in it? I tried keep= subjectId

radhikaa4 · ‎03-03-2020

I have the following dataset tthat I need to pull out or break overlap date/time HAVE: id start_date_time stop_date_time 1 3/5/2020 0:00 3/5/2020 23:12 1 3/5/2020 9:20 3/5/2020 9:45 WANT: id start_date_time stop_date_time 1 3/5/2020 0:00 3/5/2020 9:20 1 3/5/2020 9:20 3/5/2020 9:45 1 3/5/2020 9:45 3/5/2020 23:12 I am not sure how to do this and I have been using this as a source but after writing the code, it didn't work. Any help would be greately appreciated! http://support.sas.com/resources/papers/proceedings09/079-2009.pdf

radhikaa4 · ‎12-27-2019

I have the following dataset category group measure value height control mean 55 height control median 54 weight control mean 123 weight control median 125 height treatment mean 57 height treatment median 56 weight treatment mean 147 weight treatment median 167 I am trying to transpose by category and treatment then by mean/median with values Something like this: category group mean median height control 55 54 weight control 123 125 height treatment 57 56 weight treatment 147 167 I did: proc sort data=have; by category group;run; proc transpose data=have out=want; by category group; var Value; run; Not working....

radhikaa4 · ‎12-26-2019

I have following dataset and I am trying to concat. percent formatted to 9.1 (already did that) id freq percent 1 1 33.3 2 1 11.1 3 1 25 But when I use this one below: combine = cats(freq, ' ', '(', percent, '%)') There is 1) no space between freq and percent and all my percent numbers are super long , like this: id freq percent combine 1 1 33.3 1(33.3333333%) 2 1 11.1 1(11.1111111%) 3 1 25 1(25.0000000%)

radhikaa4 · ‎12-25-2019

Hi I currently have a data set where I am trying to get the proc means for all numeric variables - n, mean, median, std, min, max, and confidence intervals Here is the code I wrote: proc means data=test n mean lclm uclm alpha=0.05 median min max missing ; output out=test_stat; by group; var time1 time2 time3 time4; run; The code works in the result viewer but when i look at test_stat (it has not included upper and lower confidence intervals) Any help would be great! Thanks

radhikaa4 · ‎12-17-2019

I currently have this table below reasons_list answer frequency percent reason #1 no reason #1 yes reason #2 no reason #2 yes I would like the table below reasons_list no_frequency no_percent yes_frequency yes_percent reason #1 reason #2 I tried with proc transpose and it works but i have to do multiple transposes. Is there any way to fix with just 1-2 two steps?

Online Status	Offline
Date Last Visited	‎11-17-2020 06:53 PM

Merging two dataset to create one full

Re: Merging more than 2 datasets

Merging more than 2 datasets

min of two date columns

Re: SAS overlay histogram

date add days

Re: SAS overlay histogram

add day to datetime

SAS overlay histogram

Re: Manage Date Conflicts - Overlap

Merging two dataset to create one full

Re: Merging more than 2 datasets

Merging more than 2 datasets

min of two date columns

Re: SAS overlay histogram

date add days

Re: SAS overlay histogram

add day to datetime

SAS overlay histogram

Re: Manage Date Conflicts - Overlap

Manage Date Conflicts - Overlap

proc transpose not working

concat (catx)

proc means - data not outputed

transpose