A standard left join should work. ie proc sql; create table want as select a.id, a.value, b.product from tablea a left join tableb b on a.id=b.id; quit; You can (should) post your code if something like the above isn't working for you. ... View more

Please don't use ALL CAPS...its painful to read. Can you show the code you're trying to use? Is this code running on another computer, not through Citrix? ... View more

Confidence interval and cut offs are related so essentially the same test. I recommend a multistep model building approach. 1. Hold back a set of data, ie use 250k for model fitting and 250k as model testing 2. Use a cutoff of 0.025 or less to test if a variable is significant 3. Test in the hold back dataset for significance. 4. Repeat with different hold back samples to ensure there is actually an effect. You'll also want to differentiate between statistically significant results and practically significant. ... View more

Post the code and perhaps we can help debug rather than start from scratch? ... View more

For the entire dataset? You can do a proc transpose, but I'm guessing there's some other criteria as well? ... View more

What do you mean by dynamic variables in this context? What are you changing and what do you expect the output to be? ... View more

You can always use a  format as well, this is useful if you want to recategorize variables without really creating a new variable. *create a format; proc format; value age_fmt 12 - 14 = 'Tweens'; run; *print results; proc print data=sashelp.class; format age age_fmt.; run; *summarize by the new formats; proc means data=sashelp.class; class age; var weight height; format age age_fmt.; run; ... View more

Google inline styles and formatting. Find the unicode symbol and use that code. ie try title2 'Converting Dollars to Euros ^{unicode 0024}1 = ^{unicode 20AC}0.1794'; title3 'Converting Pounds to Yen ^{unicode 00A3}1 = ^{unicode 00A5}144.705'; title4 'You can use copyright ^{unicode 00A9} registered ^{unicode 00AE} and trademarks ^{unicode 2122} too'; ods text = "^{style [fontsize=18pt] big^{super text}} or ^{style [fontsize=10pt] small ^{sub text}}"; Stole this from a presentation at SGF, inline style formatting, Enhancements to inline formatting in SAS 9.2 is the titile but I can't find a link to the document except on SASCanada.ning.com ... View more

your probably looking for a type of array where you can do something like the following: The number of row variables will change so SAS doesn't know how many to create which will be an issue, if you have an idea of the max, set it to that. Or use an earlier dataset to create a macro variable with the largest number of = sign. Here's an example: data _null_; set edc_recipe end=eof; retain max=0; nr=countc(recipe_parameters_1, "="); if nr>max then max=nr; if end then call symput ('maxc', max); run; %put &maxc.; data test_table ; set edc_recipe; Nr = countc(RecipeParameters_1,'='); array row{&maxc}  row1-row&maxc; do L = 1 to Nr; Row(L) = scan(RecipeParameters_1,2,'='); Row(L) = substr(Row_L,1,length(Row_L) - 2); end; run; ... View more

Not as far as I know... The guessingrows option is also invalid in proc import with Excel A couple of things you might look into: 1) converting excel file to CSV and doing a proc import. Modify the SAS generated code to make sure your formats are the way you want them to be. 2) Read file with DDE 3) Can try the Force option in proc append, but not sure how that would work. If all the variable that were mismatched are missing in the new set it could be fine, but if variables that are not missing are in diff format you would lose information. 4) If all files are the same, try the option mixed=yes on all files to see if the imports are consistent. ... View more