About _MooMoo

_MooMoo · ‎04-18-2021

Hello SAS communities! I am working with time-to-event data and the main exposure is a three level categorical variable (X is either 0 or 1 or 2). I ran proc phreg and got two hazard ratio estimates which compares 1) 0 vs. 1 2) 0 vs. 2 when I am setting X=0 as reference level. I am wondering if this statistically means the same thing when I subset two data sets, one with X=0 and 1 and another with X=0 and 2. And if I run two separate cox regression, will I get the same estimate? Any advice is appreciated!

_MooMoo · ‎03-06-2021

Superb! Thanks.

_MooMoo · ‎03-06-2021

Hello SAS communities! I have following data set. id start end X 1 1/15/2021 1/21/2021 1/02/2021 2 1/22/2021 1/28/2021 1/09/2021 And I would like to output following data set. id start end X 1 1/01/2021 1/07/2021 1/02/2021 1 1/08/2021 1/14/2021 1/02/2021 1 1/15/2021 1/21/2021 1/02/2021 2 1/08/2021 1/14/2021 1/09/2021 2 1/15/2021 1/21/2021 1/09/2021 2 1/22/2021 1/28/2021 1/09/2021 The rule is.... if X is not in between start and end, create rows until X is in between start and end. For example, from the above table, rows in bold are added. Any type of advice is greatly appreciated! Thank you!

_MooMoo · ‎02-11-2021

This may be stupid question.. but I would like to generate a two way frequency table - sex by age group where age group is first stage strata. I coded.. proc surveyfreq data=x; strata age_grp; tables age_grp * sex; weight wt; run; And I got the warning saying that age_grp cannot appear both in strata and tables statement. proc surveyfreq data=x; tables age_grp * sex; weight wt; run; If I just write above code, wouldn't SAS assume there is no stratification at the first stage? Thank you!

_MooMoo · ‎02-11-2021

@ballardw Thank you for your response! The purpose of subsetting the data is because of our limited resources. Instead, we chose some portion from master data and test them (primary reason is the cost of testing all of the samples are too high). In our result, we want to take into account for the fact that our analytic data is sampled from master data.

_MooMoo · ‎02-11-2021

Hello SAS communities! I am unsure whether to use PROC FREQ with weight option or PROC SURVEYFREQ. The analytic data set I am working with is a subset of the master data set. We randomly select sample by age category from master data set and calculate weight using sampling fraction. For example, we have 1000 participants in the master data set, where 500 are below 50 and 500 are above 50. We sampled 100 participants from below 50 yrs old and sampled 50 participants from above 50 yrs old. We assigned a weight of 5 to below 50 yrs old group and assigned a weight of 10 to above 50 yrs old group. In this case, would we use PROC FREQ with weight option or PROC SURVEYFREQ with weight option and specifying age group as strata? Any advice is greatly appreciated!

_MooMoo · ‎10-10-2020

Hello! I fitted piecewise linear regression. The model specification is as follows... ,where Intervention was being in effect right after block 24. Through this model, my ultimate goal is to check whether the rate has changed from pre-intervention to post-intervention. The reason why I chose piecewise model is because I also wanted to check how rate change overtime before and after intervention. To check the ultimate goal, I ran following code... proc glm data=X; class intervention; model rate=intervention block spline; lsmeans intervention/cl; run; However, LSMEANS statement wouldn't give me the answer I want. Actually, LSMEANS statement would give me pre-intervention and post-intervention rate when block variable and spline variable were set to its mean value. This is not what I wanted. Does anyone has an idea how to output the mean rate of pre-intervention and mean rate of post-intervention from the model? Thank in advance!

_MooMoo · ‎09-21-2020

As I more think about this, I think the approach I explained above is wrong. Since time dependency of risk getting main outcome is already absorbed in the baseline hazard rate even though I did not stratify by time interval, there is no point running stratified cox regression where stratification level is time interval.

_MooMoo · ‎09-16-2020

Hello, I am wrapping my head around to figure out how to model stratified cox ph model. Assuming baseline hazard change over time, I partitioned follow-up time into two time intervals. 1 time interval for early to mid period and another time interval for late period. For example, let's say a person was followed 10 years, which can be represented as (0,10). I partitioned this time interval into two pieces such as (0, 8 ) falls into first stratum (time_strata=1) and (8,10) falls into second stratum (time_strata=2). Suppose the main covariate I am interested in is X1 and partition data set is data set I partitioned the follow-up time into two time-intervals from original data set. I coded SAS as below... proc phreg=partition; class X1; model Y=X1; strata time_strata; run; The code above gives me the exact same output as below code... proc phreg=original; class X1; model Y=X1; run; Why is this happening? Can I perform stratified analysis in this way (having time interval as strata)? If not, any other suggestions are appreciated. Thanks in advance!

_MooMoo · ‎09-16-2020

Aha! That's it.

_MooMoo · ‎09-16-2020

Thank you for the suggestion! However, when I add ref=X1 statement after param=glm, it gives an error like below. ERROR 22-322: Syntax error, expecting one of the following: FIRST, LAST. ERROR 76-322: Syntax error, statement will be ignored.

_MooMoo · ‎09-15-2020

Hello SAS communities! Suppose I would like to fit where X1 and X2 are dichotomous variables. Below code will give me HR of X1 at each level of X2 due to interaction term. proc phreg data=demo; class X1 X2 / param=ref; model (start, stop) * Y(0)=X1 X2 X1*X2; hazardratio X1/diff=ref; However, I am trying to output overall effect of X1. In other words, I would like to say - Y increased/decreased by __% when X1=1 comparing to X1=0 after accounting for other covariates in the model. In that case, will below code output the number I need to make above statment? LSMEANS only work under param=glm... will this different from when setting param=ref? proc phreg data=demo; class X1 X2 / param=glm; model (start, stop) * Y(0)=X1 X2 X1*X2; lsmeas X1/cl; Thank you! If you have other suggestions, I appreciate that too!

_MooMoo · ‎09-09-2020

Thank you for your response once again @FreelanceReinh. Could you explain further how can we use SAS functions to simulate suitable survival time? My plan was to set lambda as a baseline hazard and V as decreasing rate of hazard (e.g., hazard assumed to be decreased by 10% over time). And plug in those two parameters in the equation that article suggested.

_MooMoo · ‎09-09-2020

Hi @FreelanceReinh ! Aha, that explains why. Thank you for your clarification! Another question... If I want to generate survival time that baseline hazard rate is 0.035 and its monotone decrease rate is at 0.1, then can I set the scale parameter (lambda) as 0.035 and shape parameter (v) as 1-0.1=0.9 and plug those numbers into Bender's equation? The point I am missing is that how to set parameters in Weibull distribution and which procedure (Bender's equation or RAND function) to use in order to generate survival time? Thanks in advance!

_MooMoo · ‎09-09-2020

Hello! I am trying to generate survival time using Weibull distribution for simulation study. I got the survival time equation for Weibull distribution from Bender: https://epub.ub.uni-muenchen.de/1716/1/paper_338.pdf. That is, I am trying to confirm survival time calculated from above equation gives the somewhat similar result using RAND('WEIBULL') procedure and the sample SAS code I wrote is below. Here, I am assuming B effect is null so that exp(B'x)=1: If I am correct, both X and W would give somewhat similar answer. And when I looked at the mean of X and W value, they are quite different. Below is the output... I am unsure why this is happening. Any kinds of advice are welcomed! Thank you!

Online Status	Offline
Date Last Visited	‎04-19-2021 02:52 AM

Cox regression analysis when main exposure is a three-level categorica...

Re: Insert empty rows until certain conditions met

Insert empty rows until certain conditions met

Re: PROC SURVEYFREQ vs. PROC FREQ

Re: PROC SURVEYFREQ vs. PROC FREQ

PROC SURVEYFREQ vs. PROC FREQ

LSMEANS in time series

Re: Stratified Cox PH model: time-stratification

Stratified Cox PH model: time-stratification

Re: PROC PHREG: hazard ratio when interaction term is added

Re: PROC PHREG: hazard ratio when interaction term is added

Re: Generate survival time using Weibull distribution

Re: Generate survival time using Weibull distribution

Re: Generate survival time using Weibull distribution

Re: PROC PHREG: hazard ratio when interaction term is added

Cox regression analysis when main exposure is a three-level categorica...

Re: Insert empty rows until certain conditions met

Insert empty rows until certain conditions met

Re: PROC SURVEYFREQ vs. PROC FREQ

Re: PROC SURVEYFREQ vs. PROC FREQ

PROC SURVEYFREQ vs. PROC FREQ

LSMEANS in time series

Re: Stratified Cox PH model: time-stratification

Stratified Cox PH model: time-stratification

Re: PROC PHREG: hazard ratio when interaction term is added

Re: PROC PHREG: hazard ratio when interaction term is added

PROC PHREG: hazard ratio when interaction term is added

Re: Generate survival time using Weibull distribution

Re: Generate survival time using Weibull distribution

Generate survival time using Weibull distribution