LanMin, Whether you need to use a LEFT JOIN or not depends on your question ("requirement" as LinusH said).  I'd suggest that you read the PROC SQL documentation about left vs inner join and if you still have questions, post with samples of A, B, and WANT for us to better help you. Be careful with the "*" syntax.  Since gvkey is in both tables, you will have two variables with the same name in WANT and SAS will give unexpected results. It looks like you are using the LIBNAME approach to connecting to the database.  The only way to guarantee that the query runs on the database (rather than the SAS server) is to use a pass thru query. There are non-SQL ways to do this (with a DATA step and MERGE), but all the processing would be done in SAS, so there is no value to the UPLOAD. Doc Muhlbaier Duke ... View more

Lily, Contrary to what Steve said, I often include binary variables as continuous in analyses.  It can create some very nice shortcuts for summarizing a lot of data.  In previous versions of SAS, we had to do that in order to include the categorical predictor when there was no CLASS statement available (e.g. PROC LOGISTIC in version 6).  The binary variables had to be coded 0-1 for this to work. I think that the reason you are getting the different estimates when you treat Role as continuous is that it is coded 1 and 2.  If you recode it as a 0-1 variable, you should get the same result for the model estimation with it as a class or continuous variable.  Like Steve said, you will not be able to use some of the optional tools if it is treated continuous. Doc Muhlbaier Duke ... View more

PROC IMPORT scans the Excel records to determine the data type, it does not use the Excel format.  If you don't have EGuide, the only way I know how to convert is with the DATA step mentioned earlier. ... View more

Based on Arthur's link, you will need to use the Excel LIBNAME so you can get the NLS support to read the file.  You will not be able to use the basic EGuide import functions as that pulls in the Excel file before the NLS support can be applied.  You may not be able to do this with a Unix platform without also installing the PC Files Server. ... View more

Check out Chris H's comments on 64 bit SAS:  The top gotchas when moving to 64-bit SAS for Windows - The SAS Dummy ... View more

We may be able to provide more help if you share your code.  The DENSITY statement defaults to a normal density distribution.  If your data are skewed toward the left, that could explain why the curve goes below 0.  See http://support.sas.com/documentation/cdl/en/grstatproc/62603/HTML/default/viewer.htm#density-stmt.htm The KERNEL density estimate uses a non-parametric smoother and will follow the histogram better for non-normal data. Doc Muhlbaier Duke ... View more

Enterprise Guide can handle that conversion for you, but if you are using PROC IMPORT you will have to post-process the data. If gvkey is desired to be numeric, you can use PROC DATASETS to apply a format (Z6.) to it. If gvkey is actually a character string that happens to be all digits, then you will need to use a data step to convert, something like DATA new; SET _LAST_(RENAME=(gvkey=gvkeyold)); LENGTH gvkey $6.; gvkey=PUT(gvkeyold,Z6.); DROP gvkeyold; RUN; ... View more

One possibility is that your license for SAS on the Server has expired.  Can you open a SAS session directly (not from EGuide)? Otherwise, this is a good one to take to SAS Tech Support. Doc Muhlbaier Duke ... View more

Might still be enough.  Where you can get into sample size trouble is if your categorical variables have lots of levels as the real determinant is the number of degrees of freedom, not the number of variables per se. ... View more

Alice, 14,000 is certainly enough to build a model.  It may or may not be informative, and only the validation will help on that.  One concern with 'region' is that it may have a lot of levels, so it may degenerate for some of them. Another thing that you may be able to get to sharpen your estimates include an estimate of affluence (if you have addresses with ZIP+4, you can link that to some census files for mean income).  There is some evidence that winback campaigns work less well for those at the extremes of the socio-economic spectrum. Doc ... View more

The minimum number of predictors is 0.  With that degenerative model, the logistic regression becomes just a test of whether the outcome proportions are equal. If you have one predictor, the logistic also has equivalents from  Stat 1.  If you have a binary outcome and a classification predictor, it is a type of chi-squared test.  If the classification variable is continuous, it is roughly the same as a z-test.  The calculations are slightly different, but the results are asymptotically equivalent. Having a lot of predictors may get you a more precise model for that set of data (higher c-index), but it is not necessarily the most accurate or reproducible in general.  When you get new or additional data, the model with many variables may break down.  Harrell and others have done a lot of simulation work on the impact of over-specifying a model.  If you have lots of data (as we often do in data mining), it is less of an issue. Doc Muhlbaier Duke ... View more