rhigh's articles:

Items(347) /35 Next >> Last >|

Re: proc score #4 0 16 (3/9/2010 7:38:26 PM) comp.soft-sys.sas Hansi, You can either drop the _model_ variable when you call PROC LIFEREG: proc lifereg data=indat outest=OUTEST(drop=_model_); or rename the _model_ variable when you call PROC SCORE: proc score data=new... rhigh

Re: SAS nlmixed for count data with random effect 0 58 (3/9/2010 7:20:20 PM) comp.soft-sys.sas Nitish, It would help to know more about the data set, but with 9.2 you can do the same analyses with GLIMMIX: PROC GLIMMIX method=laplace ; * or method= quad; model count = lnW / dist=poisson solution; rando... rhigh

Re: proc phreg with large dataset 0 33 (3/2/2010 6:16:39 PM) comp.soft-sys.sas Christine, * the code you have proc phreg data=outf.fro_analysis_2_female nosummary; model (start_age,end_age)*outcome_flag(0) = birth_cohort_1 birth_cohort_2 start_age sex /ties=exact; run; is actually equi... rhigh

Re: How to specify the ESTIMATE statement in PROC MIXED #7 0 30 (2/26/2010 7:10:44 PM) comp.soft-sys.sas Susie, I'd be extra careful about running it too carefree, though one can collect output stats with multiple runs with the ODS persist option: ods listing close; ods output estimates(persist=proc)=est tests3(... rhigh

Re: How to specify the ESTIMATE statement in PROC MIXED #5 0 10 (2/26/2010 4:55:08 PM) comp.soft-sys.sas Why would you want to apply GLM when it very likely can't handle the repeated measures covariance nearly as well as MIXED or GLIMMIX? With that in mind, it is helpful to compare the new LSMESTIMATE statement w... rhigh

Re: Distribution #3 0 27 (2/24/2010 9:40:42 PM) comp.soft-sys.sas Mohsen, My first reaction is why are the counts for plots I and J so much smaller and you want to test for equality? Given that they could be smaller, you can test that the proportion of the counts for the pl... rhigh

Re: CIBASIC #4 0 12 (2/17/2010 4:12:17 PM) comp.soft-sys.sas Jules, It's not totally clear what your input data look like, but assuming they are already on a log scale and you have at least 2 observations for each level of mfgr and hrs, you could run: ods select BasicI... rhigh

Re: Model fit for ordinal multinomial models in GENMOD? 0 9 (2/12/2010 6:50:43 PM) comp.soft-sys.sas Tobias, It is somewhat of a multi-step process, namely first determine if the cumulative logit model assumption is OK with LOGISTIC: PROC LOGISTIC DATA=indat; CLASS xvar / PARAM=glm; FREQ count; MODEL y = xva... rhigh

Re: Adding/dropping a random term in the model 0 11 (2/11/2010 10:28:25 PM) comp.soft-sys.sas Bhupinder, Without further details on what you are doing and the design factors, this note is very incomplete. However, if you have version 9.2, I'd look at the COVTEST statement (experimental), something like... rhigh

Re: Inter- rater agreement for echocardiography measurements in #2 0 11 (2/11/2010 6:43:06 PM) comp.soft-sys.sas Marieta, Without further info to the contrary and based on the summary stats you sent, I'd still compute ICC under one of the two following scenarios, depending on how the raters are chosen: DATA tst(KEEP= i... rhigh

rhigh's replies:

Items(10) /1

Automating age bin creation 2 28 (2/10/2010 4:53:48 PM) comp.soft-sys.sas Dear all: I want to create agebins in months but I find myself writing several lines of codes for this task. Given the age range I have my current method will take a long time to get all the codes in. Is the... linkyox (32)

anova "type" models in proc mixed 1 25 (10/22/2009 5:06:26 PM) comp.soft-sys.sas we're trying to get some input about a simple model we've constructed to predict fish abundance....groups of a particular species of fish were collected during certain water quality conditions throughout multip... tarponquest (3)

"Non-est" issue in the Proc Mixed Procedure 1 14 (6/22/2009 9:00:00 AM) comp.soft-sys.sas Dear Friends, We encountered a problem when we tried to get the estimates with the estimate statement in the PROC MIXED procedure, SAS always indicated "Non-est" in the output window and we still do not figur... medpower (36)

Behrens-Fisher' D with Proc logistic 1 29 (6/10/2009 2:35:38 PM) comp.soft-sys.sas Hi All, To my knowledge, SAS does not provide the Behrens-Fisher' D statistics in logistic procedure. (Behrens-Fisher's D is a measure of separation for a binary performance variable. It is the difference in t... yingtaoliu (78)

2K Factorial Design Partial Confounding 2 37 (4/9/2009 8:29:03 AM) comp.soft-sys.sas I'm a graduate student using SAS for the first time. I'm able to do 2^k designs, for example a 2^3 factorial design with 2 blocks: data engineer; input speed geo angle block effect @@; datalines; 2 2 2 1 22 ... cylurian (22)

Are the "hotsolutions" postings SPAM or are they valid? 2 44 (2/25/2009 8:49:26 PM) comp.soft-sys.sas Being new on SAS-L, I'd like to know if these postings by hotsolution@msn.com are valid or are they SPAM? On Wed, Feb 25, 2009 at 9:47 AM, hotsolution wrote: Regards, John Burton Chattanooga TN ... jrburtonsaspro (103)

ANCOVA testing for homogeneity of variances 2 27 (2/5/2009 4:22:43 PM) comp.soft-sys.sas This is rather embarrassing to write about, as I feel as though I should know something as basic as this by now. I am attempting to run some very basic ANCOVA models. There is only one class variable with two t... rutz (46)

A Variable value combination Question - VERY VERY URGENT 4 22 (2/4/2009 4:16:03 PM) comp.soft-sys.sas I have the following dataset with two variables (subject, Group): subject Group ------ 10021 Perga 12124 double 12545 Perga 47845 Single I need... need_sas_help (102)

Calculating adjusted mean differences 1 27 (12/14/2008 3:54:58 PM) comp.soft-sys.sas Does anyone have any examples of SAS code about how to calculate adjusted mean differences with corresponding 95% CI from a multivariable model? (I currently have the model set up with PROC GENMOD.) I'm inter... fraeman (1)

Boxplot without whiskers ? 9 17 (6/27/2008 7:01:07 PM) comp.soft-sys.sas Hi there, Is there a trick to suppress the whisker from the boxplot generate from proc boxplot? Thanks Ya ... ya.huang (1962)