Math Forum / Math Software / SAS / February 2010

Lee,
There are a number of SAS technical notes on this.  If you're not on
9.2 phase 2, you may have to intall that to solve the problem (other
than, of course, simply opening the file in Excel and saving it in an

I want to test varaible y in regression given the independat variables
x,y,z,w,l.
need to produce a boxplot ( scatter plot ) for all at once .
Kindly help witrh SAS Code.

I get the following error when I use a .xlsx file but if I use a .xls
it works fine. Does the server need installed a new version of office?
9    proc import
10   datafile = 'test.xlsx'

I used the code -
DATA home;
    INFILE 'c:\Cwa\Home.txt';
    INPUT Owner $ 1-7 Description $ 9-33 Cost;

I'm using the code :-
data   period1;
    infile 'c:\Cwa\period1_sales_data.dat.';
    Input     TransactionID $1-8

Norton Internet Security 2010 v17.5.0.127 Final
download all antivirus programs 2010 free now
Avira AntiVir PersonalEdition 8.2.0.334
http://alagmysoft.blogspot.com/2010/02/avira-antivir-personaledition-820334.html

CATS? CATX? ||?  CAT?
Just use whichever method of concatenation you prefer, plus PUT statements
to make sure they are properly formatted.
-Joe

proc glm contrasts give me tests that assume equal variances.  if i enter
the same coefficients into spss, i get two sets of tests (equal variances
assumed and not).  can i get these tests from glm?
thanks

Can anyone comments on why SAS introduced the new function char()?
What is the advantage of char(a,1)? Isn't it the same as
substr(a,1,1)?
Thanks

Doyle Kalumbi <djrk0003@COMCAST.NET> wrote

%macro a;
  data a;
     %do i=1 %to 10;
        X&i=&i;

I have used Proc Genmod and Repeated Subject statement to obtain robust
standard errors in Poisson and negative binomial regression. Could anyone
please tell me something  about the STATEMENT AND possible SYNTAX that can be
used to obtain robust standard errors in Zero inflated ...

A simple question that appears to have no simple answer.  I am running a large number of 2sls regressions with proc syslin.  Each regression contains a number of useful but meaningless dummy variables.  I would like SAS to only print the parameter estimates for one or two key ...

You changed 2 places in the original code, which caused trouble:
set new POINT = choice NOBS = 10;
The 10 has to be n, a variable used later.
PROC REG DATA = analysis NOPRINT outest= outests;