[R] Concerning reading of SAS-files

[Frank E Harrell Jr]

bogdan romocea wrote:
> The first one is an index, not a data set. Anyway, just use SAS to
> export the data sets in text format (CSV, tab-delimited etc). You can
> then easily read those in R. (By the way, the help for read.xport says
> that 'The file must be in SAS XPORT format.' Is .sas7bdat an XPORT
> file? Hint: no.)
> 
> 
> 
>>-----Original Message-----
>>From: Fredrik Thuring [mailto:frt at Codan.dk] 
>>Sent: Friday, August 12, 2005 4:52 AM
>>To: r-help at stat.math.ethz.ch
>>Subject: [R] Concerning reading of SAS-files
>>
>>
>>
>>Hi!
>>
>>I'm trying to start a credibility estimation study with a 
>>coule of data 
>>sets that are created for SAS. The data sets are saved as 
>>".sas7bndx" and 
>>".sas7bdat".
>>I've tried reading them to R with the function 'read.xport' but this 
>>returns the error message 'Error in lookup.xport(file) : 
>>unable to open 
>>file'.
>>Are there any other functions that one could use instead?
>>
>>Thanks a lot to who ever can solve my problem!
>>
>>Fredrik Thuring
>>Codan Insurance, Copenhagen

If you want to go the CSV route, the sasxport.get function in Hmisc can 
help.  Its help function references a Howto on our web site.  Beware 
that SAS PROC EXPORT which it uses is not really a PROC but a stored 
macro (where do all the billions of $ SAS earns go?) that has a bug if 
you have an unmatched quote in a field.  sasxport.get will detect the 
problem because it compares the number of observations to that reported 
by PROC CONTENTS.

Frank Harrell