1) In general, the convergence criteria which you set with the CONVERGE= option used to stop the iteration process is R.  This is typically how PROC MODEL determines convergence; however, convergence can also be satisified if the objective function value falls below the value of the SINGULAR= option.    2) RPC will be missing when there is no previous iteration available to compute a value.  This could happen, for instance, when the initial parameter estimates meet the problem's convergence criteria.  If you're seeing a missing RPC in other cases I'd be intersted in seeing your program. ... View more

Hi, thanks for the info. I will copy my post to that forum. ... View more

Hi, there:   The model is fairly complicated. There are about 1 million observations and I checked the variance covariance matrix of the explanatories is full rank. However, the computation of RPC is missing, but everything else can show. That makes me wonder what it tells me. ... View more

Use proc logistic ... View more

Hi Jing, I spoke with the developer of proc model and he had this response, " For the SUR method MODEL computes a pseudo inverse of the OLS residuals’  covariance matrix.  When the OLS covariance matrix is singular a warning is produced explaining which equations’ rows  in the covariance matrix are linearly dependent, and PROC  MODEL continues with the SUR estimation of parameters.  If the user isn’t getting this warning message for a known deficient matrix then something else is going on " He also requested that he would be happy to look at your code and data if you wouldn't mind providing and example of the issue. - Ken ... View more

Hi. Try this... first code: and save it. %macro test; %let USER_UMACRO_PATH =F:\Community_SAS; %put &USER_UMACRO_PATH; options sasautos=("&USER_UMACRO_PATH" %sysfunc(getoption(SASAUTOS))); %mend test; Second go to sasv9.cfg file and add you autocall library path(see the last line i have modified): -SET SASAUTOS  (                 "!sasext0\core\sasmacro"                 "!sasext0\assist\sasmacro"                 "!sasext0\eis\sasmacro"                 "!sasext0\ets\sasmacro"                 "!sasext0\graph\sasmacro"                 "!sasext0\iml\sasmacro"                 "!sasext0\or\sasmacro"                 "!sasext0\qc\sasmacro"                 "!sasext0\share\sasmacro"                 "!sasext0\stat\sasmacro"   "F:\Community_SAS"                ) Now close sas editor and reopen the new one and check by invoking the macro... %test; Hope it helps.. Thanks, Shiva ... View more

Dan: That is great! Thanks. ... View more

Rick and TomTom. Many thanks for your response and help. I used Matlab pretty often before, and there are tons of sample codes of user-defined functions online that I can borrow or use as references. It is not very easy to find such resources in SAS. As my defense, even I read Rick's book, I probably don't remember all the functions used in SAS/IML. That being said, I will read all the documents recommended by both of you. Best wishes. ... View more

hi ... another idea that works with PDF output you can arrange output on a page with GOPTIONS and HSIZE, VSIZE, HORIGIN, and VORIGIN plus the STARTPAGE=NEVER option in ODS once you have the PDF output, if you can do a screen capture, you can make a bit-mapped version (e.g. PNG) proc sort data=sashelp.cars out=cars; by horsepower; run; ods pdf file='z:\plots.pdf' startpage=never notoc; ods listing close; ods results off; goptions reset=all ftext='calibri' htext=2 gunit=pct ; * no axis labels; axis1 label=none order=10 to 50 by 5 minor=(n=4); axis2 label=none order=50 to 400 by 50 minor=(n=5); * a nice looking DOT plus a spline; symbol1 f='wingdings' v='l' h=2 i=smp90 w=2; * some white space around plots; title1 ls=2; title2 a=90 ls=2; title3 a=-90 ls=2; goptions hsize=4in vsize=5in; proc gplot data=cars; goptions horigin=0in vorigin=0in; where type eq 'SUV'; footnote1 ls=2 h=3 'SUV'; plot mpg_highway*horsepower / noframe vaxis=axis1 haxis=axis2; run; goptions vorigin=5.5in; where type eq 'Truck'; footnote1 ls=2 h=3 'TRUCK'; plot mpg_highway*horsepower / noframe vaxis=axis1 haxis=axis2; run; goptions horigin=4.25in; where type eq 'Sedan'; footnote1 ls=2 h=3 'SEDAN'; plot mpg_highway*horsepower / noframe vaxis=axis1 haxis=axis2; run; quit; goptions vorigin=0in; footnote1 ls=2; proc gslide; note h=5 j=c 'DATA SET SASHELP.CARS' j=c h=3 j=c 'PLOTS OF MPG(HIGHWAY) VS HORSEPOWER' j=c '<more stuff>' ; run; quit; ods pdf close; ods listing; ods results; ... View more