tulip; Import code below is what I use to import. Below that is the code I use to export the same file once I've run it through some SAS steps. PROC IMPORT OUT= WORK.NewOrdFaults DATATABLE= "NewOrderFaults" DBMS=ACCESS REPLACE; DATABASE="\\Moftp\MainFrameFtp\ESADOWN\NewOrderFaults.MDB"; SCANMEMO=YES; USEDATE=YES; SCANTIME=NO; RUN; PROC EXPORT DATA= WORK.UpDater OUTTABLE= "NewOrderFaults" DBMS=ACCESS REPLACE; DATABASE="\\MOFtp\MainFrameFtp\EsaDown\NewOrderFaults.MDB"; RUN; ... View more

code that works that may give you some direction ... PROC IMPORT OUT= WORK.NewOrdFaults DATATABLE= "NewOrderFaults" DBMS=ACCESS REPLACE; DATABASE="\\Moftp\MainFrameFtp\ESADOWN\NewOrderFaults.MDB"; SCANMEMO=YES; USEDATE=YES; SCANTIME=NO; RUN; ... View more

Celine; You will need to produce the 6 individual graphs to the approximate size required (use hsize & vsize options) into a catalog (default is gseg). You then need to replay these 6 charts into the final rtf file using proc greplay. hth ... View more

I surmise that you are using a vbar chart. I often use hbar charts instead because they provide a lot more space for labels and such. ... View more

My understanding is that ms needs a datetime field. This is what I've used. In your case, I think you will have to add time 00:00:00 to the sas date and then go through the steps below. Hopefully this will get you started. Asof=sasdate plus sastime as a SAS datetime value;**Build sas datetime value; **Build microsoft date serial; AS_Of=(int(asof/86400) + 21916) + ((asof-(int(asof/86400)*86400))/86400); ... View more

I don't know about alternate columns necessarily, but here's what I used to shade specific columns. For readability, every third or even fourth column is actually quite sufficient. First, build a colour format with something like Proc format: value $bkgrnd "Lvl1','Lvl2','Lvl3'='cxf6f5ea' other='cxffffff'; run; Then in proc tabulate, under the ODS umbrella, apply the $bkgrnd format to the class variable that contains the values of Lvl1, Lvl2,Lvl3,... with classlev LvlVarname/style=[background=$bkgrnd.]; Not my invention, I think Cynthia helped me through this one. You should be able to find more in the archives in the ODS section. Hope this helps at least a little bit! ... View more

Peter; The most convincing method is to show the pie and then ask data questions. Then show the other (for different data so they can't rememember the answers) and note the change in understanding, response time etc. I've sometimes shown pie charts without any values attached along side the comparable bar chart and asked for responses to data questions. ... View more

Take a look at this paper ... http://www.perceptualedge.com/articles/visual_business_intelligence/save_the_pies_for_dessert.pdf ... View more

I'm able to send mail to firstname.lastname@gmail.com with these options: -emailhost=usenet.xxxxxx.ca -emailsys=SMTP where "xxxxx" also happens to be the name of our domain. ... View more

Eva; The axis statement that I posted earlier (sorry that it was so terse) pre-supposes that the dates will be SAS date values (i.e. days since 01Jan1960). To force quarters to be the first day of the quarter, I often use the intnx function as follows: QtrDate=intnx("Quarter',ActualDateValue,0,'b'); This results in values like 01JanYY, 01AprYY,01JulYY,01OctYY when formatted as date5. This date field will then plot very nicely with a qtr increment in the axis statement. Hope this helps, Bill ... View more

indeed there is axis1 order=("&strtyr"d to "&Grfend"d by qtr); ... View more

Avi; I think your data might be better displayed with a vbar graph in proc gchart. Bar graphs are better for detecting values, line graphs are better for detecting trends. I expect that gchart will provide the axis break more naturally. Be sure to show a note somewhere indicating that null periods are not shown. wd ... View more

go to support.sas and search for ERROR: Connect: The Microsoft Jet database engine cannot open the file One of the responses may well apply to your situation. ... View more

Here's one way ... data Test; input Name $1-4 Number Value; datalines; Jim 3 4 Jim 4 5 Jim 6 7 Jim 6 8 Jim 3 2 Suzy 3 1 Suzy 2 1 ; run; proc freq data=Test; table Name/out=SuprvCnt; run; proc sort data=test; by name; run; data Test2; merge Test (in=a) SuprvCnt (in=b where=(Count >4)); by Name; if b; run; ... View more