Hi,   Is there a way of using stepwise to select variables for multinomial logistic regression?    Thank you! ... View more

  Hello,   I have two questions about COMPARE function.    1. I am looking at this link (https://go.documentation.sas.com/?docsetId=lefunctionsref&docsetTarget=p0iaa25pnhl59ln1v4yrympngpic.htm&docsetVersion=3.1&locale=en) and I am not clear about the definition of 'name literal.'  Does anyone have more examples that explain the term?    n or N removes quotation marks from any argument that is a name literal and ignores the case of string-1 and string-2. A name literal is a name token that is expressed as a string within quotation marks, followed by the uppercase or lowercase letter n. Name literals enable you to use special characters (including blanks) that are not otherwise allowed in SAS data set or variable names. For COMPARE to recognize a string as a name literal, the first character must be a quotation mark.   2. Example 1 the link has the code below and one of the sample string comparison, '$12'n $12 has three different cases to show the result of COMPARE.  I am not clear how the result with a modifier 'nl' (#14 in the table below)did not get the 0 result.   Thanks!       data test; infile datalines missover; input string1 $char8. string2 $char8. modifiers $char8.; result=compare(string1, string2, modifiers); datalines; 1234567812345678 123 abc abc abx xyz abcdef aBc abc aBc AbC i abc abc abc abc l abc abx abc abx l ABC 'abc'n ABC 'abc'n n '$12'n $12 n '$12'n $12 nl '$12'n $12 ln ; proc print data=test; run;   Results of Comparing Two Strings Using the COMPARE Function ... View more

Hi,   Does anyone know how intercept is calculated in PROC REG or PROC SURVEYERG when weight is assigned?  Is it calculated as a weighted mean of outcome?  Or is weighted ignored?   Thank you, Momi ... View more

Hi, How can I adjust the size of the regression output in lst or text file so that I have all solutions are lined up in one row?   This is what I have so far...                                   Solution for Random Effects                                                        Std Err Effect              Estimate       Pred     DF   t Value A                      0.1081      0.01360   36E4      7.94 B                      0.02681    0.01810   36E4      1.48 C                      0.03403    0.01608   36E4      2.12                                    Effect              Pr > |t| A                           <.0001 B                            0.1387 C                            0.0344   I want the p-value output to be next to the t value.(below)                               Solution for Random Effects                                                        Std Err Effect              Estimate       Pred     DF   t Value        Pr > |t| A                      0.1081      0.01360   36E4      7.94    <.0001 B                      0.02681    0.01810   36E4      1.48    0.1387 C                      0.03403    0.01608   36E4      2.12       0.0344     Do I need to add some options in a SAS program?  Thanks!   ... View more

  Hi, How can I leave a zero count when outputting the count from PROC FREQ? For example, here is the result from PROC FREQ.     Yes No A 40 0 B 30 20 C 0 30   When I use an OUT option, the output only contains nonzero count (below).   A Yes 40 B Yes 30 B No 20 C No 30   Instead above, I want to have all counts including zeros (below).   A Yes 40 A No 0 B Yes 30 B No 20 C Yes 0 C No 30   Can anyone help on this?   Thanks!   ... View more

Hi I've been use this _null_ statement to go to a new slide in ODS PowerPOint.  However, some papges do not work...   data _null_ ; dcl odsout obj(); run;     The code below overlaps in one slide even though there is the _null_ statement.  Could anyone help on this? Am I missing something?   Momi     ods layout gridded x = 0 pct y = 25 pct; ods region; proc odstext; P ' '; P 'ods graphics / width = 4.5 in height = 4 in;'; P 'ods layout gridded columns = 2 x = 0 pct y = 10 pct;' ; P '^{super }ods region; /* Graph 1 */' ; P '^{super }proc sgplot --- for Male'; P '^{super }ods region; /* Graph 2 */' ; P '^{super }proc sgplot --- for Female'; P 'ods layout end;'; run; ods layout end; data _null_ ; dcl odsout obj(); run; title; footnote; ods layout gridded columns = 2 ; ods graphics / width = 4.5 in height = 4 in; title1 "2 Graphs Per Slide"; ods region; /* Graph 1 */ proc sgplot data = sashelp.class; where sex = "M"; scatter x = height y = weight ; title "Male: Height vs. Weight"; run; ods region; /* Graph 2 */ proc sgplot data = sashelp.class; where sex = "F"; scatter x = height y = weight ; title "Male: Height vs. Weight"; run; ods layout end;   ... View more

Does anyone know how to add a space in ods text?     I want to do something like below in Proc ODSTEXT   hello      hello           hello   I tried to use the code below,,,    proc odstext;          P "Hello";          P "     Hello"  ;          P "           Hello";         run;   But this would give me   Hello Hello Hello   Pelase let me know if you know how to do this.  I want tp do this on PowerPoint.   Thanks,   Momi     ... View more

Does anyone know how to do the pairwise comparison for Fisher's exact test?   Thanks,   Momi ... View more