Providing software solutions since 1976
49838 - TWOSAMPLESURVIVAL statement options EVENTSPERGROUP=, GROUPEVENTS=, and EVENTSTOTAL= produce incorrect results- Problem Note
In SAS 9.3 TS1M0 and SAS 9.3 TS1M1, the EVENTSPERGROUP= and GROUPEVENTS= options in the TWOSAMPLESURVIVAL statement in PROC POWER produce incorrect results. In addition, the power computed from the EVENTSTOTAL= option ...
49888 - The OUTSTAT= data set in PROC CALIS includes an observation where _TYPE_=STD- Problem Note
The documentation says that for the default COV analysis in PROC CALIS, _TYPE_=STD is not included in the OUTSTAT= data set, but in fact it is. You might want to take out this observation so PROC SCORE can be used ...
48940 - Conjugacy Sampler in PROC MCMC produces incorrect results with MISSING=CC- Problem Note
When the prior and posterior distributions are of the same family of distributions (called a conjugate prior) and the MISSING=CC option is specified, PROC MCMC produces incorrect results. There is currently no ...
48896 - Predicted values in PROC FMM with missing values might be wrong- Problem Note
If you have missing values in the input data set and use the OUTPUT statement to obtain predicted values, then some of those predicted values might be incorrect. This problem happens intermittently, so the only ...
49703 - Results may be incorrect when STRATA and POSTSTRATA variables are the same- Problem Note
The results produced by PROC SURVEYMEANS might be incorrect when any of the variables specified in the STRATA statement are also specified in the POSTSTRATA statement. To avoid this problem, create one or more ...
48111 - PROC SURVEYMEANS may produce incorrect variance estimates with poststratification- Problem Note
The variance estimates and related statistics are wrong in most cases when the POSTSTRATA statement is specified in PROC SURVEYMEANS. A Hot Fix addresses this issue.
49581 - MIN and MAX values displayed in the INSETGROUP may be incorrect- Problem Note
If you specify the MIN and MAX keywords in an INSETGROUP statement in PROC BOXPLOT, the minimum and maximum whisker values are displayed when ODS Graphics is enabled. This means if you have specified BOXSTYLE=SCHEMATIC, ...
49413 - ODDSRATIO statement reports nonestimable ("Non-est") values- Problem Note
If the PARAM=GLM option is specified on the CLASS statement in PROC LOGISTIC, and the REF= option is also used to change the reference level of a CLASS variable from the default, the odds ratios estimates produced by ...
49211 - PROC GLIMMIX PLOTS=BOXPLOT does not create display for random intercept models- Problem Note
If you specify a random intercepts model using the SUBJECT= effect syntax and request PLOTS=BOXPLOT, the GLIMMIX procedure will fail to generate the ODS GRAPHICS BOXPLOTS. For example, proc glimmix data=test plots= ...
48329 - PROC SURVEYREG will abend or produce missing values with VARMETHOD= , DOMAIN and estimation or testing statements- Problem Note
If the option VARMETHOD=JACKKNIFE|JK or VARMETHOD=BRR is specified on the PROC SURVEYREG statement, a DOMAIN statement is specified, and any of the following statements are also specified, then the procedure will abend ...
48922 - Incorrect log likelihood, type1, type3 tests for ZIP model when WEIGHT statement is specified- Problem Note
When a zero-inflated Poisson (ZIP) model is requested in PROC GENMOD and the WEIGHT statement is specified, then the Log Likelihood, Full Log Likelihood, AIC, AICC, and BIC statistics in the Criteria For Assessing ...
48541 - Abnormal termination if ROC PRED= and BY statements are used together- Problem Note
In PROC LOGISTIC, if you specify both a BY statement and an ROC statement that uses the PRED= option then the procedure will terminate abnormally. To avoid the problem, specify the PRED= variable in both the MODEL and ...
48517 - Utility file error may occur with large data sets and the SURVIVAL= option on the OUTPUT statement- Problem Note
A utility file error, such as
48518 - Incorrect area standard errors with NOFIT option, ROC, and BY statements- Problem Note
If you specify both the BY statement and one or more ROC statements as well as the NOFIT option in the MODEL statement, then the standard error of the area under the ROC curve reported by ROC statements is incorrect. To ...
48514 - GENERAL function in PROC MCMC produces incorrect density estimates- Problem Note
When the GENERAL function is used on the RANDOM statement in Proc MCMC it produces an incorrect density estimate.
48513 - FMM does not reset model selection limits during BY group processing- Problem Note
When model selection is attempted for BY group processing FMM only behaves correctly for the first BY group
48299 - Scoring of observations with missing response not done for some binary response models in PROC GENMOD- Problem Note
If your response is binary and you specify the DIST=BINOMIAL option in the MODEL statement, and if you also specify a CLASS statement, then observations that are missing only on the response are not scored by the OUTPUT ...
48270 - Abend or incorrect variance estimates if replicate contains all missing values- Problem Note
PROC SURVEYMEANS will abend with errors or it will complete with some incorrect results with VARMETHOD=JK or BRR, DOMAIN or BY and REPWEIGHTS.
48286 - Incorrect predicted values with SELECTION=, full-rank parameterization, and missing values- Problem Note
If you use the SELECTION= option in the MODEL statement, and the PARAM=REF, PARAM=EFFECT, or PARAM=ORDINAL option in the CLASS statement, and if there are missing values in some variables that are not included in the ...
38449 - Read Access Violation error in PROC BOXPLOT or PROC SHEWHART- Problem Note
When the HAXIS= or VAXIS= option is used on the PLOT statement in PROC BOXPLOT to assign an axis definition and the corresponding AXIS statement contains a text description suboption after quoted text instead of before ...
48223 - The TEST statement in PROC HPMIXED might generate an exception error- Problem Note
The TEST statement in PROC HPMIXED might generate an exception error message. A hot fix is planned for this issue.
47834 - PROC GLIMMIX stops subsequent processing in batch mode if one BY-group fails to converge- Problem Note
When using the BY statement in PROC GLIMMIX and a BY-group fails to converge, PROC GLIMMIX terminates and does not process the remaining groups in the input data set. This only happens when the program is submitted in ...
48120 - Possible incorrect ESTIMATE statement results for zero-inflated models- Problem Note
If an ESTIMATE statement involving zero-inflation parameters precedes another ESTIMATE statement involving parameters of the mean model, the results of the statement involving the mean model will be incorrect. To avoid ...
48045 - Proc MCMC exits unexpectedly- Problem Note
Proc MCMC will exit without any explanation when the hyperparameters of a random effect are in the functional form of a regression model.
48031 - Graph produced with PLOTS=PROBPLOT option in PROC LIFEREG is incorrect- Problem Note
The ODS Graphics plot produced by the PLOTS=PROBPLOT option on the PROC LIFEREG statement is incorrect. The data to compute the Kaplan-Meier estimates (points on the plot) are not sorted first. The probabilty plots ...
