SAS Support Communities

Even though it looks like a non-Windows system, the path to the STP directory is awfully long (121 characters): (/opt/app/SAS/work/SAS_work87D2000036D6_esessux0015/SAS_work8E07000036D6_esessux0015/7e539000-ab52-11e1-8001-313430333831/) I know for a fact that copying SAS software depot into Myfiles directory or into Desktop, causes PC SAS installation process to fail. I don't know how your STP directory gets assigned and whether this is relevant at all in your case, but if you have any input into this assignment, try shortening this path. ... View more

What is your LIB library? Is it defined using the ODBC engine? Perhaps it has something to do with the problem reported in this SAS note for Access to Oracle: SAS/ACCESS ® software's LIBNAME engines support creating tables as well as reading from and writing to tables, but they do not support table altering. http://support.sas.com/kb/37/015.html ... View more

Hi Neetu, I think you wanted the days to be on the X axis and finish time on the Y axis. If not, you need to swap the order in the plot statement. You can comment out the goptions changing the size of the chart, font, etc. I had to generate some data. Try this code whether it works for you. data a; format finish_time TIMEAMPM11. dy ddmmyy8.; finish_time='04:55:23't; do dy='18JUL12'd to '25JUL12'd by 1;   finish_time=finish_time  120*ranuni(0) + 10*ranuni(0);   output;   end; run; options nocenter nodate byline; goptions reset=all ftext="Albany AMT" htext=0.35cm ctext=BLACK fontres=presentation border vsize=4.2in hsize=7.5in noprompt NOGSFPROMPT; SYMBOL1      INTERPOL=JOIN      HEIGHT=10pt      VALUE=NONE      LINE=1      WIDTH=2      CV = _STYLE_; Axis1      STYLE=1      WIDTH=1      MINOR=NONE      LABEL=( "TIME" ); Axis2      STYLE=1      WIDTH=1      MINOR=NONE      LABEL=( "DATE" ); TITLE; TITLE2 "Line Plot"; PROC GPLOT DATA = a; PLOT finish_time*dy / VAXIS=AXIS1 HAXIS=AXIS2 FRAME; RUN; QUIT; TITLE2; FOOTNOTE; GOPTIONS RESET = all; Each time you generate a plot it is appended to the gseg catalog as Gplot, Gplot1, Gplot2, Gplot3, etc.. You must press page Down to see the last chart. When you want to delete all old plots and start from scratch, submit the following code: /* delete previously created grsegs before creating new ones */ proc greplay igout=work.gseg nofs;    delete _all_; run; quit; ... View more

Hi Shri, If you did not find a correct answer here is one more tip: data have2; format f1 f2 $30. fn1 8.2 fn2 tot tot2 8.2 ; input f1 1-11 f2 12-22; fn1=input(substr(f1,1,length(compress(f1))),best8.); fn2=input(substr(f2,1,length(compress(f2))),best8.); tot=sum(0,fn1,fn2); * if you have many fni columns you may use an array instead; array f  {*} fn1 -- fn2; tot2=0; do i=1 to dim(f); tot2=sum(0,tot2,f); end; drop i; cards; 1          229.00        2          39.83         3          38.25         4                        1          346.00                   26.75         1          55.64                                  ; run; options nocenter; proc print; run; *Output: Obs         fn1         fn2         tot        tot2 1         1.00      229.00      230.00      230.00 2         2.00       39.83       41.83       41.83 3         3.00       38.25       41.25       41.25 4         4.00         .          4.00        4.00 5         1.00      346.00      347.00      347.00 6          .         26.75       26.75       26.75 7         1.00       55.64       56.64       56.64 8          .           .          0.00        0.00 ; Hope this helps, Dorota ... View more

Read this SAS Note: http://support.sas.com/kb/17/008.html or an expanded tutorial here: SAS programming guide, sas error, Catalog SASHelp.Dmine does not exist Hope this helps, Dorota Message was edited by: Dorota Jarosz - added a link to expanded tutorial ... View more

Shri, (actually this was an attempt to answer RV's question) I am getting an error with the length-1 third argument and I should.  What is even more odd, I am getting single digit hundredths values rather than integers. I guess I have to study the input function :smileyshocked: data have; format f1 f2 $30. f1n 8.2 f2n 8.2 ; input f1 1-11 f2 12-22; f1n=input(substr(f1,1,length(trim(f1))),8.2); f2n=input(substr(f2,1,length(trim(f2))),8.2); cards; 1          229.00        2          39.83         3          38.25         1          346.00        1          55.64         ; run; options nocenter; proc print; run; *Output: f1n values in decimals? Obs    f1                                f2                                     f1n         f2n 1     1                                 229.00                                0.01      229.00 2     2                                 39.83                                 0.02       39.83 3     3                                 38.25                                 0.03       38.25 4     1                                 346.00                                0.01      346.00 5     1                                 55.64                                 0.01       55.64 ; data have2; format f1 f2 $30. f1n 8.2 f2n 8.2 ; input f1 1-11 f2 12-22; f1n=input(substr(f1,1,length(compress(f1))),best8.); f2n=input(substr(f2,1,length(compress(f2))),best8.); cards; 1          229.00        2          39.83         3          38.25         1          346.00        1          55.64         ; run; options nocenter; proc print; run; *Output: Obs    f1                                f2                                     f1n         f2n 1     1                                 229.00                                1.00      229.00 2     2                                 39.83                                 2.00       39.83 3     3                                 38.25                                 3.00       38.25 4     1                                 346.00                                1.00      346.00 5     1                                 55.64                                 1.00       55.64 ; Message was edited by: Dorota Jarosz - added (actually this was an attempt to answer RV's question) to the greetings. ... View more

To make your input function flexible, use the length function as the second parameter: f1n = input(substr(f1,1,length(f1),8.2); Please provide more info. How are you importing the csv file? Do you use the missover option in the filename statement? ... View more

SAS assigns the best16. format, if none is specified. Try adding a format statement for the new variables. format f1n f2n 8.2; f1n=sum(0,input(f1,8.2); By the way, do you really need the character variables to be 30 characters long?  Unless you know you have very long strings (which are supposed to be numbers anyway), you should make these fields shorter. ... View more

Proc surveymeans outputs the mean, st.error, and confidence limits for the mean by default. If you are requesting other statistcs, you need to provide the specific syntax and options you are using. The interpretation of the output may depend on the position of the option in the procedure syntax. Proc statement options, Ratio statement options, etc. Standard error is the estimate of standard deviation of the estimator of the statistic in question, based on the given sample size. So yes, the confidence limits for the statistic (e.g., the mean) will be based on this standard error. Standard deviation is usually referring to the variation of the original variable raw values in the population, not the variation of a statistic from the sample with a fixed sample size. ... View more

This is also true for independent Windows workstation SAS installation trying to connect to an ODBC system database. We had to add servers to tnsnames.ora file on each PC. We also had to define this system database in Windows Data Sources (ODBC) tool. ... View more

Hi Zetter, You can output the results of proc freq to a dataset, one table at a time.  The categorization of the library visits variable can be accomplished either by hard coding or by assigning a user defined format. The following two proc freq steps produce the same output: proc freq data=library noprint; tables behaviour*lib_vis_cat /out=beh_vis_cat (drop=percent); *title "Behaviour by Library visit count - hard coded"; run; proc format; value lib_vis_fmt low-9 = " 1-9" 10-20 = "10-20" 21-30 = "21-30" 31-high="31 plus"; run; proc freq data=library noprint; tables behaviour*lib_visits / out=beh_vis (drop=percent) ; *title "Behaviour by Library visit count using format"; format lib_visits lib_vis_fmt.; run; *Both tables contain four variables: behaviour, lib_visits (or lib_vis_cat), COUNT, and PERCENT, which can be dropped using the output data set option shown. You can use this syntax for any variable in the data set, but if you include more than one cross tabultion in the tables statement, only the last one will be output.; proc print data=beh_vis noobs; title "Behaviour by library visit count using format"; run; *Output: Behaviour by library visit count using format              lib_ behaviour    visits     COUNT   bad         1-9         3   bad        10-20        2   bad        31 plus      1   good        1-9         3   good       10-20        2   good       31 plus      1   nice        1-9         1   nice       21-30        3 ; ... View more