About JJ_83

JJ_83 · ‎02-12-2021

Thank you! That's exactly what I needed!

JJ_83 · ‎02-11-2021

The problem I see with the week and month functions is that it returns the week number for the year, but since I have multiple years I need a unique number for each week. Here is what my data looks like now.

JJ_83 · ‎02-11-2021

1. The week number of 01JAN2001 and 01JAN2002 should be different. 01JAN2002 should have a week number of 53, and so on. Here's what my data looks like. 2. I already have values for date_1 and date_2, and what I want to do is match the date_1 and date_2 values to the week number for the week that it falls in. For example a date_1 value of 1JAN2001 would have a week value of 1 and a date_2 value of 31DEC2001 would have a week value of 52.

JJ_83 · ‎02-11-2021

First, I want to assign a week number to each week within the range: January 1, 2001 - December 31, 2020 Then I want to assign two date variables (date_01 and date_02) to their corresponding week number. I want to do #1 and #2 for months instead of weeks. Any ideas on how to do this easily? I am guessing there is a way to do this in one step instead of 2.

JJ_83 · ‎12-04-2020

Yes, I posted because I have not been able to find any answers through the documentation.

JJ_83 · ‎12-03-2020

I am working with a matched case control dataset and I am running the following proc logistic procedure. proc logistic descending data = data; strata newgroup; model new_case = exp; run; newgroup is the matching variable. new_case is the binary outcome variable (1/0). exp is the binary exposure variable (1/0). I am trying to replicate results that my coworker did in stata. So my first question is: How can I run logistic regression in sas with robust or cluster standard errors? What kind of standard errors does proc logistic use? (I am assuming it is not robust or cluster) My apologies if these are really stupid questions!

JJ_83 · ‎08-21-2020

Hello, I want to assign the same start and end date by matching variable. Any advice? data have; input id $ matching $ start :mmddyy10. end :mmddyy10.; format start mmddyy10. end mmddyy10.; datalines; B0000362004 A 01/01/2001 11/11/2011 B0005852005 A . . B0009422005 B 02/02/2002 12/12/2012 B1370912010 B . . ; run; data want; input id $ matching $ start :mmddyy10. end :mmddyy10.; format start mmddyy10. end mmddyy10.; datalines; B0000362004 A 01/01/2001 11/11/2011 B0005852005 A 01/01/2001 11/11/2011 B0009422005 B 02/02/2002 12/12/2012 B1370912010 B 02/02/2002 12/12/2012 ; run;

JJ_83 · ‎08-14-2020

Ahh I see, there was a typo in the day variable. I wanted the first 5 days in 2000, but I put the first 5 days in 2010. Everything works great now, thank you!

JJ_83 · ‎08-14-2020

The day variable refers to days, I just truncated it to the first 5 days in 2000. The code is not working for me.

JJ_83 · ‎08-14-2020

Given this: data have; input WELL_API $ ID $ DAY :mmddyy10. EXPOSURE_START :mmddyy10. EXPOSURE_END :mmddyy10. PHASE :best.; format DAY mmddyy10. EXPOSURE_START mmddyy10. EXPOSURE_END mmddyy10.; datalines; 01133244410000 B1392272014 01/01/2010 01/01/2000 01/05/2000 0 01133244410000 B1392272014 01/02/2010 01/01/2000 01/05/2000 1 01133244410000 B1392272014 01/03/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/04/2010 01/01/2000 01/05/2000 3 01133244410000 B1392272014 01/05/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/06/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/07/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/08/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/09/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/10/2010 01/01/2000 01/05/2000 2 ; run; I want this: All entries where Phase = (2 or 3) Exposure_start <= Day <= Exposure end data want; input WELL_API $ ID $ DAY :mmddyy10. EXPOSURE_START :mmddyy10. EXPOSURE_END :mmddyy10. PHASE best.; format DAY mmddyy10. EXPOSURE_START mmddyy10. EXPOSURE_END mmddyy10.; datalines; 01133244410000 B1392272014 01/03/2010 01/01/2000 01/05/2000 2 01133244410000 B1392272014 01/04/2010 01/01/2000 01/05/2000 3 01133244410000 B1392272014 01/05/2010 01/01/2000 01/05/2000 2 ; run;

JJ_83 · ‎08-14-2020

Hi Paige, I see what you're saying about the benefits of a long vs wide dataset. Thank you for the suggestion! I have reorganized my dataset, and was hoping you could help with my last hurdle. I want all WELL_API per ID where PHASE = 1 or 2 for any day between EXPOSURE_START and EXPOSURE_END. I'm stuck on how to do it with a long dataset. data have; input WELL_API $14. ID $11. +4 DAY :mmddyy10. EXPOSURE_START :mmddyy10. EXPOSURE_END :mmddyy10. PHASE $1.; format DAY mmddyy10. EXPOSURE_START mmddyy10. EXPOSURE_END mmddyy10.; datalines; 01133244410000 B1392272014 01/01/2010 01/01/2000 12/23/2014 0 01133244410000 B1392272014 01/02/2010 01/01/2000 12/23/2014 1 01133244410000 B1392272014 01/03/2010 01/01/2000 12/23/2014 2 01133244410000 B1392272014 01/04/2010 01/01/2000 12/23/2014 3 01133244410000 B1392272014 01/05/2010 01/01/2000 12/23/2014 0 ; run;

JJ_83 · ‎08-13-2020

I am stuck on how to create a new dataset based on a few variables I have now. What I have now is this: data have; input ID DATE :mmddyy10. PERIOD_01 PERIOD_02 PERIOD_03; FORMAT ID z14. DATE mmddyy10.; datalines; 01133244410000 01/03/2000 0 0 2 02019220960000 02/05/2000 0 2 2 07167297020000 03/15/2000 2 2 2 17067210480000 04/02/2000 0 0 0 ; run; I want the dataset structure below where DAY_01 = January 1, 2000 and DAY_N = December 31, 2015. And I want to incorporate two levels of coding: Level One: DAY_01 - DAY_31 = PERIOD_01 DAY_32 - DAY_60 = PERIOD_02 DAY_61 - DAY_91 = PERIOD_03 Level Two: All days prior to DATE = 0 The day corresponding to DATE = 1 All days after DATE = 1 until the first day that = 2 data want; input ID DAY_01 DAY_02 DAY_03 DAY_04 DAY_N; FORMAT ID z14.; datalines; 01133244410000 0 1 1 2 2 02019220960000 0 0 1 2 2 07167297020000 1 1 2 2 0 17067210480000 1 2 0 0 0 ; run; Any help would be really appreciated!

JJ_83 · ‎07-28-2020

Wow, this works perfectly! Thank you so much!

JJ_83 · ‎07-28-2020

Hello, I have a dataset with the following structure. What I want to do is keep only those observations with a day_01 to day_05 value of "1" for any day between the exposure_start and exposure_end. I would like to do this for each Participant_ID. The exposure_start and exposure_end values are unique to each participant_id Any ideas? Here is the code for this sample dataset: data have; input WELL_ID DAY_01 DAY_02 DAY_03 DAY_04 DAY_05 EXPOSURE_START $ EXPOSURE_END $ PARTICIPANT_ID; FORMAT WELL_ID z14.; datalines; 01133244410000 0 0 1 0 0 DAY_01 DAY_05 1 02019220960000 0 0 0 0 0 DAY_01 DAY_05 1 07167297020000 1 0 0 0 0 DAY_01 DAY_05 1 17067210480000 0 0 0 0 0 DAY_01 DAY_05 1 34000000000000 0 0 0 0 0 DAY_01 DAY_05 1 34001200010000 0 0 0 0 0 DAY_02 DAY_04 2 34001200020000 0 0 0 0 0 DAY_02 DAY_04 2 34001200030000 0 0 0 1 0 DAY_02 DAY_04 2 34001200040000 0 0 0 0 0 DAY_02 DAY_04 2 34001200050000 0 0 0 0 0 DAY_02 DAY_04 2 ; run;

JJ_83 · ‎07-28-2020

Thank you, this worked great!

Online Status	Offline
Date Last Visited	‎03-04-2025 03:11 PM

Re: Merging By Variable

Re: Merging By Variable

Re: Merging By Variable

Re: Merging By Variable

Merging By Variable

Re: Help Merging Datasets

Help Merging Datasets

Dataset Structure - Pairs of Observations

Very Large Dataset - Out of Resources

Re: Wide to Long Dataset

Re: Help Merging Datasets

Re: Help Merging Datasets

Re: Dataset Structure - Pairs of Observations

Re: Wide to Long Dataset

Re: Wide to Long Dataset

Merging By Variable

Re: Merging By Variable

Re: Merging By Variable

Re: Merging By Variable

Re: Week and Month Intervals

Re: Week and Month Intervals

Re: Week and Month Intervals

Week and Month Intervals

Re: Proc Logistic Standard Errors

Proc Logistic Standard Errors

Assigning values by group

Re: Conditional Creation of New Dataset

Re: Conditional Creation of New Dataset

Re: Conditional Creation of New Dataset

Re: Conditional Creation of New Dataset

Conditional Creation of New Dataset

Re: Observations Selection

Observations Selection

Re: Creating New Variables With Dynamic Suffix