cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
MiraKr_
‎03-28-2022
  • 26 Posts
  • 4 Likes Given
  • 0 Solutions
  • 1 Likes Received
About
Mira Kr. is a SAS Certified Base Programmer. She is currently working as an analyst/research coordinator at the Clinical and Health Informatics Research Group at McGill University in Montreal, CANADA. Mira Kr. is an aspiring epidemiologist involved in outcomes research of healthcare practices and interventions. She has 1- 2 years of experience with SAS . She uses SAS for accessing and managing epidemiological data (PROC IMPORT, DATA Step, PROC SQL), for performing queries and analyses (PROC FREQ, PROC MEANS, PROC GENMOD, PROC GLM, PROC GEOCODE) and to generate reports (PROC TABULATE, PROC REPORT).

Re: Kaplan-Meier Survival Plotting Macro %NEWSURV

by Obsidian | Level 7 in SAS Communities Library
‎03-28-2022 07:38 PM
‎03-28-2022 07:38 PM
Hi @JeffMeyers ,   I am looking for a macro that can handle  pre-calculated weights in order to produce a weighted HR and weighted KM (i.e. weighted median and N at risk). The weights are stored in the original dataset as variable sw. In addition, I need to add option COVSANDWICH when estimating the 95%CI. I was wondering, if you have a version of the %NEWSURV macro that can do this.   PROC PHREG DATA=data COVSANDWICH; weight sw; CLASS arm (Ref=first); MODEL OS*CNSR(1)=arm /rl=wald ties=efron; run; proc lifetest data=data; time OS*CNSR(1); strata arm; weight sw; run; ... View more

Transposing blocks of observations

by Obsidian | Level 7 in SAS Programming
‎10-02-2020 10:35 AM
‎10-02-2020 10:35 AM
Hello, I trying to convert table "Have" into table "Want". I have multiple tables "Have" with different number of steps in them and looking for an automated way to determine the number of steps. Table "Have"       step Variable Estimate Odd_Ratio 0 INTERCEPT 2.2168 9.178 (2.62, 32.149) 0 PRIOR_CSA -0.6381 0.528 (0.236, 1.18) 0 MALE -0.4896 0.613 (0.319, 1.176) 0 EASI0141 -0.00831 0.992 (0.966, 1.018) 0 AAGE 0.017 1.017 (0.993, 1.042) 0 EASI0142 0.00382 1.004 (0.96, 1.05) 1 INTERCEPT 2.2368 9.363 (2.737, 32.037) 1 PRIOR_CSA -0.6275 0.534 (0.241, 1.182) 1 MALE -0.4894 0.613 (0.32, 1.176) 1 EASI0141 -0.00638 0.994 (0.98, 1.008) 1 AAGE 0.017 1.017 (0.993, 1.042) 2 INTERCEPT 1.9037 6.711 (2.531, 17.794) 2 PRIOR_CSA -0.7005 0.496 (0.228, 1.081) 2 MALE -0.5303 0.588 (0.309, 1.122) 2 AAGE 0.018 1.018 (0.994, 1.043)   Desired format: Table "Want"                 step Variable Estimate_0 Odd_Ratio_0 Estimate_1 Odd_Ratio_1 Estimate_2 Odd_Ratio_2 ProbChiSq_2 0 INTERCEPT 2.2168 9.178 (2.62, 32.149) 2.2368 9.363 (2.737, 32.037) 1.9037 6.711 (2.531, 17.794) 0.0001 0 PRIOR_CSA -0.6381 0.528 (0.236, 1.18) -0.6275 0.534 (0.241, 1.182) -0.7005 0.496 (0.228, 1.081) 0.0776 0 MALE -0.4896 0.613 (0.319, 1.176) -0.4894 0.613 (0.32, 1.176) -0.5303 0.588 (0.309, 1.122) 0.1072 0 EASI0141 -0.00831 0.992 (0.966, 1.018) -0.00638 0.994 (0.98, 1.008) 0.018 1.018 (0.994, 1.043) 0.1364 0 AAGE 0.017 1.017 (0.993, 1.042) 0.017 1.017 (0.993, 1.042)       0 EASI0142 0.00382 1.004 (0.96, 1.05)           ... View more

Proc Tabulate N(%)

by Obsidian | Level 7 in ODS and Base Reporting
‎08-23-2018 03:50 PM
‎08-23-2018 03:50 PM
Hello, I am looking for a way to request proc tabulate to display the output with N(%) in the same column. Is that possible using proc tabulate or do I need proc template?   proc tabulate data=dquest.pooled_dri_quest missing; class STUDYID sex ; tables (sex all='Overall' ), STUDYID*(n='N'*F=6.0 colpctn='%'*F=6.1) (all='Pooled')*(N='N'*F=6.0 COLPCTN='%'*F=6.1); run;   Output:   Study Identifier Pooled DRI12544 EFC13579 N % N % N % Sex 490 63.1 1197 62.9 1687 63 F M 286 36.9 705 37.1 991 37 Overall 776 100 1902 100 2678 100   I want this:   Study Identifier Pooled DRI12544 EFC13579 N (%) N (%) N (%) Sex 490 (63.1) 1197 (62.9) 1687 (63.0) F M 286 (36.9) 705 (37.1) 991 (37.0) Overall 776 (100) 1902 (100) 2678 (100)   Many Thanks, MK ... View more

Re: How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-11-2018 11:01 AM
‎07-11-2018 11:01 AM
Thank you for the explanation, Astounding!  ... View more

Re: How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-11-2018 10:58 AM
‎07-11-2018 10:58 AM
Thank you, Tom. This solution works very well and produces identical results to Astounding's code and also flags the 0s (instead of assigning missing).  Would you mind to explain what this part of the code does: months(1+mod(month(startdt)+i-1,12)) Thank you, MK  ... View more

Re: How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-10-2018 03:35 PM
‎07-10-2018 03:35 PM
Astounding, would you mind to add some comments to this part of your code to help me understand what these statements do:    do k=0 to min(11, n_months); mm = starting_month + k; if mm > 12 then mm = mm - 12; flags{mm} = 1; end;   I understand the rest of the code.    Many thanks! MK   ... View more

Re: How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-10-2018 03:12 PM
‎07-10-2018 03:12 PM
Hi --, This solution works very well! 🙂   Thank you so much!   MK ... View more

Re: How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-10-2018 02:21 PM
‎07-10-2018 02:21 PM
The year is not important. So,  January=1 is any January.  Also, If a patient was present in Jan 2016 and Jan 2017, January=1.  Thank you, Mira ... View more

How to determine if a month falls b/w two dates?

by Obsidian | Level 7 in SAS Programming
‎07-10-2018 02:07 PM
‎07-10-2018 02:07 PM
Hello, I am trying the determine participation by month. Each patient has as a start and an end date in the dataset. I would like to check if the patient was present in Jan, Feb, Mar,...Dec. At the end, I need variables for Jan--Dec with values of 0/1 for each patient. The study spans b/w 2015-05-21 --- 2017-07-29.    I would like to avoid writing a long code that looks like this: if STARTDT le ('01JAN2016'd) le ENDDT then January=1; else if STARTDT le ('02JAN2016'd) le ENDDT then January=1; .... else if STARTDT le ('31JAN2016'd) le ENDDT then January=1;   Thank you for your insights regarding a more efficient way to do this!   MK ... View more

Re: Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-10-2017 09:40 AM
‎04-10-2017 09:40 AM
Thank you! ... View more

Re: Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-10-2017 09:39 AM
‎04-10-2017 09:39 AM
Thank you ... View more

Re: Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-10-2017 09:36 AM
‎04-10-2017 09:36 AM
This solution is very short and easy to run, it converted all 182 csv datasets into sas datasets. Thank you! ... View more

Re: Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-05-2017 11:27 AM
‎04-05-2017 11:27 AM
Hi Patrick, Thank you for the code! It works like magic. The only thing I would like to edit would be to assign the original CSV dataset names (listed in Dirlist) to the newly created SAS datasets, rather than creating SAS datasets with names dataset_01, dataset_02, etc, which is not really informative. I will try to work on that, but if you have an easy solution, please let me know.    Thank you all for posting solutions.   MK ... View more

Re: Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-04-2017 03:48 PM
‎04-04-2017 03:48 PM
Hi--, I have 50 individual files with their own unique structure and names in one folder (please see attached photo). Each uque datasets have different variables in it. How can we set up the SAS code to repeat the import procedure 50 or so times?    Regards, MK ... View more

Importing entire library of CSV datasets into SAS

by Obsidian | Level 7 in SAS Programming
‎04-04-2017 03:23 PM
‎04-04-2017 03:23 PM
Hello, I have a library of around 50 CSV datasets. I would like to import them all in one step (if possible) rather than importing individual files one by one with proc import.  Can anyone help me with this?   PS: In other words, I am trying to avoid doing this 50 times: PROC IMPORT OUT= WORK.dataset1 DATAFILE= "Path\dataset1.csv" DBMS=CSV REPLACE; GETNAMES=YES; DATAROW=2; RUN;   Thank you, MK ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎03-28-2022 11:03 PM
Likes given to
View All
Likes from
View All
Quick links