About psh23

A_Kh · ‎02-05-2024

Interesting.. This time the code produced PNG file (with higher resolution than it is produced usually with IMAGEFMT=PNG option). The same WARNING message in the log (WARNING 1-322: Assuming the symbol PS was misspelled as psepsf.)

amcr729 · ‎03-01-2022

thank you! (the problem doesn't arise when I just use complete case analysis, but I take your point about whether or not MI makes sense given amount of missing data).

psh23 · ‎04-19-2019

Thanks! valuesformat=4.2 does not work but valuesformat=data does. proc sgplot data=week noborder; /*format HazardRatio 4.2;*/ scatter x=week y=HazardRatio / yerrorlower=WaldLower datalabelattrs=(size=12pt color=black) yerrorupper=WaldUpper errorbarattrs=(color=black) markerattrs=(symbol=CircleFilled size=6 color=black); refline 1 / axis=y lineattrs=(pattern=shortdash) transparency=.5; yaxis valuesformat=data label="Hazard Ratio " min=.96 max=1.06 type=log logbase=10 valueattrs=(size=14pt color=black family="Times New Roman") labelattrs=(size=14pt color=black family="Times New Roman") ; xaxis label="Week of Gestation" min=1 max=37 valueattrs=(size=14pt color=black family="Times New Roman") labelattrs=(size=14pt color=black family="Times New Roman") minor; run;

StatDave · ‎11-30-2018

Check your data to see how many observations there are in the largest cluster (level of REGISTRY). Or, if you get partial output, see if the GEE Model Information table is there which will show the Correlation Matrix Dimension. The appearance of SAS freezing is probably actually due to it having to take a lot of time computationally because of a large cluster size. The time needed to estimate and display all of the correlations in a large matrix could be considerable. If you can reduce the cluster size, then that will reduce the time needed. If you intend to use only the independence or exchangeable structure, then you don't need the CORRW option. The single correlation estimate for the exchangeable structure is printed by default without the CORRW option. You could also try fitting the same model using the newer PROC GEE instead of GENMOD.

Rick_SAS · ‎07-10-2018

Ah! Sorry that I misunderstood you. Yes, the EFFECTPLOT statement is for visualizing a single model. In many cases, putting the BY variable into the CLASS statement produces compatible models for linear models. If so, you might be able to use the EFFECTPLOT statement anyway: Add volume1 to the model and use SLICEBY=volume1. However, there is definitely a need to check whether the predictive models are the same. Personally, if volume1 is a BY variable, I'd use PROC SGPANEL or ODS LAYOUT GRIDDED to put the plots side-by-side.

Reeza · ‎06-13-2018

99% of the time the error is with the hand calculation. Do it with a data step so you can't get any typo's or rounding errors and post your code if it still doesn't match.

TomKari · ‎06-05-2018

Assuming you mean double quotes, not parentheses, data have; CharString = '{"I34.0":1,"I48.0'; output; run; data want; retain _RX1; if _n_ = 1 then _RX1 = prxparse('|"([^"]+)"|'); set have; if prxmatch(_RX1, CharString) then WantedString = prxposn(_RX1, 1, CharString); run; Tom

psh23 · ‎03-06-2018

Yes thank you!

Ksharp · ‎03-01-2018

data have; input id date :mmddyy10. med ; format date mmddyy10.; datalines; 1 4/30/2015 2 1 5/6/2015 2 1 8/31/2015 1 1 11/30/2015 1 1 3/8/2016 2 1 6/7/2016 2 1 9/20/2016 2 2 3/26/2015 2 2 8/25/2015 2 2 11/24/2015 3 2 2/23/2016 3 2 10/3/2016 2 3 4/7/2016 1 3 5/19/2016 1 ; data temp; set have; by id med notsorted; if first.med then group+1; run; proc sql; select distinct id,min(date) as start format=mmddyy10., max(date) as end format=mmddyy10., med from temp group by group; quit;

Ksharp · ‎11-29-2017

I agreed with Rick. 242,000 observations is too much for PROC GAM to analyze. PROC GAM required the expanse system resource to run Non parameter regression. If you want fit non-linear effect, why not try decision tree PROC HPSPLIT ?

ChrisBrooks · ‎10-27-2017

My point exactly - SAS isn't going to use it if it's commented out......

Astounding · ‎11-30-2016

I'm still guessing, but see if this comes close: data want; totcount=0; do until (last.sex); set have; by Countrytract Sex notsorted; totcount + count; end; do until (last.sex); set have; by Countrytract Sex notsorted; if totcount >= 10 then output; end; drop totcount; run; The top DO loop sums up COUNT for a set of SEX/COUNTYTRACT. The bottom DO loop reads the same observations, and outputs them if TOTCOUNT is 10 or more.

psh23 · ‎11-22-2016

There are certain census tract / county strata where there are no values for a certain agegroup, because it is such a small subgroup, but those stratum are not included in the data. It seems like it is getting stuck in that it wants a value for every single age group stratum within each census tract / count. It is impossible to include these strata in the dataset because the denominator is zero and it will not run if there are zeros.

Online Status	Offline
Date Last Visited	‎07-08-2020 02:43 PM

Re: EPS output from sgplot

Re: EPS output from sgplot

Re: EPS output from sgplot

EPS output from sgplot

Re: SGPLOT yaxis format values

SGPLOT yaxis format values

Re: genmod not printing corrw

genmod not printing corrw

Re: overlay graph of predicted probabilities from two glimmix models

Re: overlay graph of predicted probabilities from two glimmix models

Re: EPS output from sgplot

Re: EPS output from sgplot

Re: PROC MIANALYZE freezes SAS

Re: SGPLOT yaxis format values

Re: genmod not printing corrw

Re: overlay graph of predicted probabilities from two glimmix models

Re: hazard ratio from regression coefficients in an interaction term

Re: cleaning character string

Re: calculate inverse weighted cumulative average across preceding wee...

Re: time varying exposure with unevenly spaced time intervals

Re: PROC GAM error

Re: sas not recognizing formatted values when creating new variable

Re: Deleting rows that have less than ten observations by two vars, af...

Re: Proc STDRATE error message, missing matched stratum