hi ... I'm never quite sure what I'll get with MIDPOINTS, that said ... I'm using V9.3 and you can add the RANGE option to show the actual low-high values (results of using MIDPOINTS) in the legend, otherwise you get the midpoints using the data set and graphics options from my previous post ... title h=3 'LEGEND WITH RANGES (BASED ON MIDPOINTS 2.5e5 1e6 2.75e6 4e6  8e6)' ls=2; proc gmap map=maps.us data=uspop; id state; choro pop / midpoints=(2.5e5 1e6 2.75e6 4e6  8e6) range coutline=grayaa; format pop comma10.; run; quit; ... View more

Hi actually, you only have to list STATE (I edited my posting !!!). ... View more

hi ... have you tried ... choro People_Count / levels=5 annotate=P; for example ... * not a total state population ... just add city populations within states; proc summary data=maps.uscity nway; var pop; class state; output out=uspop (drop=_:) sum=; run; goptions reset=all ftext='calibri' htext=1.75 gunit=pct; pattern1 v=ms c=cxe4e9fc; pattern2 v=ms c=cx7793f1; pattern3 v=ms c=cx4068ec; pattern4 v=ms c=cx1133a3; pattern5 v=ms c=cx0b226d; proc gmap map=maps.us data=uspop; id state; choro pop / levels=5 coutline=grayaa; format pop comma10.; run; quit; ... View more

hi ... since no one else has asked, do you have to use SQL ... data one; retain state; set sashelp.zipcode (keep=state city statename--dst); run; ... View more

hi ... the '--' works as a data set option, so (no NOPRINT needed when you create a table, no table alias A needed either) ... proc sql; create table one as select * from sashelp.zipcode (keep=state city statename--dst); quit; ... View more

hi ... you could also use a PUT statement as follows ... data have; input id clm_aud_nbr : $12. @@; cards; 1 111111111101 2 211111111101 3 111111111101 4 211111111101 5 311111111101 ; proc sql; create table want as select id, put(clm_aud_nbr,$10.) as claimnum from have; quit; one advantage is that using PUT gives CLAIMNUM a length of 10 using SUBSTR gives CLAIMNUM a length of 12 (same length as CLM_AUD_NBR) ... View more

hi ... well there are 37 points plotted and 37 tick marks, so what are the values for the variable M (is that a SAS date or just 1 through 37) did you try it without using the haxis=axis1 to see what happens ... View more

Hi ... 1st argument is a seed, remaining arguments assign 1, 2, 3, 4, etc with the probability you specify ... a/ probabilities do not sum to 1, RANTBL sums those specified and subtracts them from 1 to get one more assignment rule b/ probability sum  > 1, RANTBL adjusts/ignores stated probabilities data test; do _n_ = 1 to 1e5;    x = rantbl(0, 0.3, 0.4, 0.25);    y = rantbl(0, 0.3, 0.4, 0.25,  0.05);    z = rantbl(0, 0.3, 0.4, 0.35,  0.35);    output; end; run;                               Cumulative    Cumulative x    Frequency     Percent     Frequency      Percent 1       30118       30.12         30118        30.12 2       39952       39.95         70070        70.07 3       24898       24.90         94968        94.97 4        5032        5.03        100000       100.00                               Cumulative    Cumulative y    Frequency     Percent     Frequency      Percent 1       30010       30.01         30010        30.01 2       39996       40.00         70006        70.01 3       25022       25.02         95028        95.03 4        4972        4.97        100000       100.00                               Cumulative    Cumulative z    Frequency     Percent     Frequency      Percent 1       30136       30.14         30136        30.14 2       39908       39.91         70044        70.04 3       29956       29.96        100000       100.00 Another use is to randomly assign a value from an array to a variable ... array names(5) $6 _temporary_ ('mike' 'linlin' 'ksharp' 'art' 'peterc'); name = names(rantbl(0, 0.2, 0.2, 0.2, 0.2));                                    Cumulative    Cumulative name      Frequency     Percent     Frequency      Percent art          19869       19.87         19869        19.87 ksharp       19903       19.90         39772        39.77 linlin       20122       20.12         59894        59.89 mike         20089       20.09         79983        79.98 peterc       20017       20.02        100000       100.00 FYI ... there's also CALL RANTBL, and from online doc ... The CALL RANTBL routine gives greater control of the seed and random number streams than does the RANTBL function. ... View more