About trungdungtran

trungdungtran · ‎05-23-2019

@Rick_SAS: Thank you for your reply and sorry for a late response from me. I agree with you that I have a small sample size as the contingency table is matrix(data=c(17, 31, 5, 1), nrow=2, ncol=2, byrow=FALSE). However, the numbers in my first post are the exact number rather than asymptotic approximation. As you see that I take p-value from the first table (last row) and the exact confidence limits from the last two rows of the last table (see attachment please). They are all exact as I understand correctly from the output of SAS. Please see here an article talking about this problem. https://academic.oup.com/biostatistics/article/11/2/373/267728

Kurt_Bremser · ‎11-26-2018

The goal has to be to reduce the process to the least necessary steps, and to make it error-proof in the sense that the process detects problems at the earliest possible moment. The way to do this is to use a stable and reliable data transfer format (which implicitly disqualifies Excel files), and data steps that are custom-written to the file specification and will throw an error when unexpected data (unexpected either through content or format) is fed into them. All my ETL jobs follow this guideline, and the only problems that "slip through" are of a semantic type where completely plausible data is given that later is found to be logically defective.

trungdungtran · ‎11-26-2018

Thank you so much! I have implemented for a task as you have said. I did SAS and R portion independently until everything is fine then I integrate them and then I can only type in SAS and run R behind the scene.

trungdungtran · ‎11-10-2018

I agree with you that it seems to lazy. I had this feeling from the first post and now you share the same feeling.

trungdungtran · ‎11-07-2018

Thank you so much for letting me know a new function in SAS.

ballardw · ‎11-05-2018

@trungdungtran wrote: Hi all, I have a longitudinal data where each subject has a number of repeated measurements, i.e., occupying several rows in the data table. Unfortunately, the ID is not starting from 1. Now I would like to create a new ID, starting from 1. I have completed a SAS code as follows: data original; input ID ; cards; 12 12 12 16 16 19 19 19 27 27 27 ; run; data newID; set original; retain newID 1 previousID 12; if ID=previousID then newID=newID; else newID=newID+1; previousID=ID; drop previousID; run; Proc print data=newID noobs; run; This code works as I want. However, I feel the code complicated. Do you have any suggestion to make it simpler? Thank you for reading! Something like this maybe? data have; input ID ; cards; 12 12 12 16 16 19 19 19 27 27 27 ; run; proc sort data=have; by id; run; data want; set have; by id; if first.id then Newid+1; run; I explicitly sorted the data by ID just in case and so that all of the values of ID end up with the same newid.

trungdungtran · ‎11-01-2018

Thank you @FreelanceReinh for suggestion about the blog post. I am learning macro so I take this as an exercise for me to practice.

trungdungtran · ‎11-01-2018

Yes, the option "reset=index" solves the problem! Thank you @PGStats.

Rick_SAS · ‎03-15-2018

For information about how to store any table or graph to a data set, see ODS OUTPUT: Store any statistic created by any SAS procedure For the specific case of graphs, see How to get data values out of ODS graphics

Online Status	Offline
Date Last Visited	‎05-27-2019 01:50 PM

Re: Fisher's exact test: p-value or 95% confidence limits

Fisher's exact test: p-value or 95% confidence limits

Re: Advantage of calling R into SAS

Re: Reading Excel file with mixture of date formats

Reading Excel file with mixture of date formats

Re: Calculate percentage between the variables

Re: Calculate percentage between the variables

Re: Advantage of calling R into SAS

Re: Advantage of calling R into SAS

Advantage of calling R into SAS

Re: Advantage of calling R into SAS

Re: Calculate percentage between the variables

Re: How to compute time distance for longitudinal data in SAS

Re: Advantage of calling R into SAS

Re: Fisher's exact test: p-value or 95% confidence limits

Re: Reading Excel file with mixture of date formats

Re: Advantage of calling R into SAS

Re: Calculate percentage between the variables

Re: How to compute time distance for longitudinal data in SAS

Re: Create new ID starting from 1

Re: How to get confidence interval for root MSE in proc reg

Re: Each macro submission, new plots with different names are created

Re: Save output for proc PRINCOMP