About tamhane

tamhane · ‎10-24-2018

Haven't come across that in SAS but PASS software has it. The reference they cite is: Wu, Jianrong. 2015. “Sample size calculation for the one-sample log-rank test,” Pharmaceutical Statistics, Volume 14, pages 26-33. Also attached is the documentation from PASS and a poster.

tamhane · ‎10-24-2018

SAS does report standard error with it. Uno's Concordance Statistic Source Estimate Standard Error Model 0.5464 0.1584 For 95% CI: you would multiple SE with 1.96. In the given example, it would be 0.1584*1.96 = 0.3105 Lower limit = 0.5464 - 0.3105 = 0.2359 Upper limit = 0.5464 + 0.3105 = 0.8569 Thus if rounded 0.55 (0.24, 0.86) or expressed as % then 54.6 (23.6, 85.7). Hope this helps. Ashutosh

tamhane · ‎10-24-2018

Here are some papers 1. Using SASÒ to Match Cases for Case Control Studies by Kawabata et al. 2. Using SAS® Software to Perform a Case-Control Match on Propensity Score in an Observational Study by Parsons. 3. SAS® Programs to Select Controls for Matched Case-Control Studies by Matthews and Brill 4. Matching Data Using SAS! Programming and Human Judgment by Go. 5. Simplified Matched Case-Control Sampling using PROC SURVEYSELECT by Disekar 6. Incidence-Density Sampling: Alternative Programming Approaches to Achieve the Same End by Gately and Tumbull 7. Automating the Selection of Controls in Case-Control Studies by Mounib 8. An Introduction to Matching and its Application using SAS® by Mandrekar

tamhane · ‎10-22-2018

For correlated data, you would use ID statement. See: https://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_phreg_sect056.htm When the proportional hazards assumption is not met, you would use STRATA statement. See: https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_phreg_sect019.htm and https://www.cibmtr.org/meetings/materials/crpdmc/documents/2010/feb2010/loganb_coxproportion1.pdf (page 7)

tamhane · ‎10-18-2018

Examining the same variable at 2 different points is a paired design. If continuous variable, then you can do T1-T2 (or T2-T1) to get the differences and then run PROC UNIVARIATE to get the p-value for Signed rank test (Null hypothesis is that the difference is equal to zero). If the variable is categorical then it would be a 2x2 paired table and can use McNemar's test to get the p-value which compares the marginal frequencies. The odds ratio here would be labelled as conditional odds ratio that will be calculated using the discordant cells (Null hypothesis is OR=1). Hope this helps. Ashutosh

tamhane · ‎10-17-2018

Just to highlight the results of the NIE and % mediated. macro Causalmed p-value 0.047 0.055 % mediated 28.78 15.96 Thanks.

tamhane · ‎10-17-2018

Try using PROC SGPLOT on the output data.

tamhane · ‎10-17-2018

Hi, I was comparing the results of the PROC CAUSALMED procedure and the SAS macro given in the paper "Mediation Analysis Allowing for Exposure–Mediator Interactions and Causal Interpretation: Theoretical Assumptions and Implementation With SAS and SPSS Macros" by Valeri and VanderWeele. Outcome=Binary Exposure/Treatment=Binary Mediator=Binary All coded as 0 and 1. However, the results obtained somewhat differ. A] Program and Output of the macro %mediation(data=anak23,yvar=BPControlBoth,avar=VL_LT200,mvar=MMAS_NonAdh,cvar=,a0=0,a1=1,m=0,yreg=logistic,mreg=logistic, interaction=false,casecontrol=,output=,c=,boot=,cens=); run; SAS Output Obs Effect Estimate p_value _95__CI_lower _95__CI_upper1234 cde 2.25450 0.044196 1.02132 4.97666 nde 2.25450 0.044196 1.02132 4.97666 nie 1.22482 0.047073 1.00262 1.49627 total effect 2.76137 0.013251 1.23613 6.16857 SAS Output Obs Effect Estimate1 proportion mediated 0.28777 B] Program and Output of the CAUSALMED proc causalmed data=anak23; model BPControlBoth = VL_LT200 MMAS_NonAdh; mediator MMAS_NonAdh = VL_LT200; run; SAS Output The CAUSALMED Procedure Summary of Effects Estimate StandardError Wald 95%Confidence Limits Z Pr > |Z|Total EffectControlled Direct Effect (CDE)Natural Direct Effect (NDE)Natural Indirect Effect (NIE)Percentage MediatedPercentage Due to InteractionPercentage Eliminated 0.2299 0.0913 0.05089 0.4089 2.52 0.0118 0.1932 0.0908 0.01530 0.3711 2.13 0.0333 0.1932 0.0908 0.01530 0.3711 2.13 0.0333 0.0367 0.0191 -0.00078 0.07417 1.92 0.0550 15.9613 9.7584 -3.1648 35.0875 1.64 0.1019 0 . . . . . 15.9613 9.7584 -3.1648 35.0875 1.64 0.1019 Thanks. Ashutosh

tamhane · ‎10-17-2018

Refer to this paper: "Mediation Analysis Allowing for Exposure–Mediator Interactions and Causal Interpretation: Theoretical Assumptions and Implementation With SAS and SPSS Macros" by Valeri and VanderWeele.

tamhane · ‎01-19-2018

Thanks much. Ashutosh

tamhane · ‎12-13-2017

I was reading the (attached) article entitled "Log-Link Regression Models for Ordinal Responses" by Blizzard et al published in Open Journal of Statistics, 2013 , 3, 16-25. Wondering if anyone has SAS syntax for that. Thank you. Ashutosh

Online Status	Offline
Date Last Visited	‎11-08-2018 07:08 PM

Re: Sample size calculation for a one sample logrank test

Re: How to obtain the 95% confidence interval using Uno's concordance ...

Re: Case:Control Matching with Conditions

Re: Correlated Data with PHREG?

Re: How to get P values for the same variable at 2 points in time.

Re: Mediation: Causalmed vs Macro

Re: How do I change the title size in a graph (proc lifetest)?

Mediation: Causalmed vs Macro

Re: Can CAUSALMED procedure handle more than one mediator variables?

Re: Log-Link Regression Models for Ordinal Responses