About cminard

cminard · ‎04-08-2020

This does not always work, and is not the recommended method. Please see the Gibbs paper.

cminard · ‎04-08-2020

Thanks for posting. I had a similar problem and was able to use this information to fix it. My solution was a little different, but the steps I followed were: 1. Deleted the UserChoice registry key as discussed in the Double-clicking paper. Mine was located here: HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\FileExts\.sas\UserChoice 2. Running the SASFileTypeRegFix.reg registry fix as discussed in the paper. downloaded and ran that file. 3. Using the windows registry editor. (Modify at your own risk, take a back-up before modifying): Browse to HKEY_CLASSES_ROOT\SAS.Program.701\shell\OpenwithSAS940\command Change value to: "C:\Program Files\SASHome\SASFoundation\9.4\core\sasexe\sasoact.exe" action=Open datatype=SASFile filename="%1" progid=SAS.Application.940 Browse to HKEY_CLASSES_ROOT\SAS.Program.701\shell\Open\command Change value to: "C:\Program Files\SASHome\SASFoundation\9.4\core\sasexe\sasoact.exe" action=Open datatype=SASFile filename="%1" progid=SAS.Application.940

cminard · ‎04-06-2020

Time-to-event models are probably fine. I think that you have too many age group categories. In my opinion, you can use the results of this analysis to identify better age groupings. That is, which age groups can you combine together? You could also leave age as a continuous variable in the CPH regression model. I would NOT delete the observations for age groups with no events. I would reduce the number of age groups by combining similar groups. Also, I personally would not define 10-year age groups. I would let the data analysis determine the categories. For example, identify the quartiles (25th, 50th, and 75th percentiles), then create 4 age groups based on those quartiles. Use KM analysis to see if the curves are different between the 4 age groups. Then combine similar age groups.

cminard · ‎04-05-2020

proc freq data=<dsn>; tables race; run;

cminard · ‎03-29-2020

The purpose of the survival analysis is to account for censoring and to use all available data. I suggest reporting the log-rank p-value for the KM curve. Then, the plot can include 95% confidence bands. If you're interested in reporting, say, the 1- and 2-year marks, then you can report the survival probabilities with 95% confidence intervals by group and time point. I don't think that it's advantageous or useful to do a formal hypothesis test at those time points.

cminard · ‎11-05-2019

data temp; input ID VisitDate:mmddyy10.; format VisitDate mmddyy10.; datalines; 6960 8/30/2016 11073 6/8/2016 11093 8/23/2016 11129 9/28/2017 11274 4/13/2016 11274 10/10/2016 11274 5/8/2017 11288 11/30/2017 11315 9/25/2017 11315 2/27/2018 11315 8/21/2018 11322 8/8/2017 11453 1/27/2017 11453 8/15/2017 11453 1/29/2018 11453 7/17/2018 11533 4/14/2016 11533 9/7/2017 11533 6/13/2018 ; run; proc sort data=temp; by id visitdate; run; data temp2; set temp; by id; if first.id then VisitNumber=1; else VisitNumber+1; run;

cminard · ‎11-05-2019

http://theobligatescientist.blogspot.com/2010/02/r-i-still-love-you-but-i-hate-your.html

cminard · ‎10-23-2019

An increase of 1 pill prescribed is associated with a 1.033 fold increased odds of consuming 1-10 pills compared with no pills. An increase of 1 pill prescribed is associated with a 1.053 fold increased odds of consuming 10+ pills compared with no pills. But this analysis seems a bit odd practically speaking. Maybe agreement (ie, kappa)?

cminard · ‎05-30-2019

Sounds like a doubly repeated measures design. Maybe take a look at this document: https://support.sas.com/resources/papers/proceedings15/SAS1919-2015.pdf

cminard · ‎05-15-2019

Can you just use sgplot? proc sgplot data=_; format t yymmn6.; series x=t y=x; xaxis min=1004; run;

cminard · ‎05-15-2019

I think that it should just be "subject=Animal". Probably won't make a difference though. This line is telling GLIMMIX that you have repeated measures within each animal. it will estimate the correlation matrix between locations. Make sure the data is sorted by animal id then location.

cminard · ‎05-13-2019

I think that it should be "subject=Animal". You have locations nested within each animal. So, the measurements within each animal are going to be correlated. I don't think that subject=Animal(TRT) is correct, but it may not make a difference depending how you coded the variables. You specify type=cs which mean a compound symmetric correlation structure. That is, you are assuming the same correlation between all of the locations. That's probably giving you equal SEs. You could try type=un which will estimate different correlations for all pairs of locations.

cminard · ‎02-15-2019

Perhaps this example is relevant to your situation: https://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_phreg_sect056.htm

cminard · ‎02-14-2019

how about something like this: data foodtypes; length food type $ 12; input food $ type $; datalines; apple fruits apple fruits apple fruits apple fruits apple vegetables apple fruits artichoke vegetables artichoke vegetables asparagus vegetables asparagus vegetables broccoli vegetables broccoli vegetables carrots vegetables eggplant vegetables eggplant fruits orange fruits orange fruits orange fruits orange fruits orange fruits orange fruits pear vegetables pear fruits pear fruits pear fruits pear fruits pear fruits pepper vegetables pepper fruits ; run; proc freq data=foodtypes noprint; tables food*type / out=crosstab; run; proc freq data=crosstab noprint; tables food / out=foodfreq; run; data discrepant; set foodfreq; if count=2; run; proc sql noprint; select "'"||strip(food)||"'" into :FoodList separated by ',' from discrepant; quit; proc print data=crosstab noobs; var food type; where food in (&foodlist); run;

cminard · ‎11-15-2018

Please see the following paper: Hsieh, F. Y., and Lavori, P. W. (2000). “Sample-Size Calculations for the Cox Proportional Hazards Regression Model with Nonbinary Covariates.” Controlled Clinical Trials 21:552–560. In this paper, the authors state the following: "In a regression model, the variance of the estimate b 1 of the parameter θ 1 is inversely related to the variance of the corresponding covariate X 1 ." Therefore, the variance of the parameter estimate would get smaller as the variance of the covariate increases. I have not yet worked through the reasoning behind this argument though.

Online Status	Offline
Date Last Visited	2 weeks ago

Re: Running R in SAS

Re: Running R in SAS

Re: Running R in SAS

Running R in SAS

Re: SAS9.4 freezes/non responding when copying

Re: SAS9.4 freezes/non responding when copying

Re: ods word with table of contents using proc(s)

Re: More than 2 Repeated variables in PROC MIXED

Re: Using array to count observations per group

Re: conditional combinations

Re: Running R in SAS

Re: Running R in SAS

Re: macro variable list elements

Re: macro variable list elements

Re: PROC FREQ McNemar test

Re: SAS9.4 freezes/non responding when copying

Re: SAS9.4 freezes/non responding when copying

Re: Count successive missing values

macro variable list elements

Re: What is the correct SAS code for a mixed linear model with a rando...

Re: Fixing .SAS file associations with SAS 9.4

Re: Fixing .SAS file associations with SAS 9.4

Re: Cox proportional HM - missing data

Re: Number of variables

Re: Kaplan-Meier - different p-values according to different time poin...

Re: Creating visit number for longitudinal data

Re: An oddity Round function:- Can somebody who has used both R and SA...

Re: interpreting multinomial logistic regression

Re: repeated repeated measures analysis

Re: Gaps Before and After GPLOT

Re: Glimmix repeated LSmeans

Re: Glimmix repeated LSmeans

Re: Appropriate test for clustered survival data

Re: Rename duplicates and fish out the odds

Re: COXREG results using PROC POWER