Home Page for the TeradataForum
 
 

 

Archives of the TeradataForum





Message Posted: Fri, 03 Dec 2004 @ 15:51:55 GMT





     
  <Prev Next>   <<First <Prev
Next>
Last>>
 




Subj:   Re: Best way of writing SQL to retrieve a sample
 
From:   Michael Larkins



Although a function like RANK will work for this, it is not a random sample. Instead, it will always return the highest values or optional the lowest values. Plus, you did not tell us which release of Teradata you are running, so here a way to get a random sampling in both V2R4 and V2R5:

     sel country_code, state_code, county_code
     from (sel country_code, state_code, county_code, csum(1,1) as randomizer
            from country_table   group by 1  qualify randomizer < 51) DerivedTable

If you have V2R5 and prefer ANSI standard SQL, you should be able to do the same thing with ROW_NUMBER.


Hope this helps,

Michael Larkins
Certified Teradata Master
Certified Teradata SQL Instructor





     
  <Prev Next>   <<First <Prev
Next>
Last>>
 
 
 
 
 
 
 
 
 
  
 
  Top Home Join Privacy Feedback  
 
 
Copyright for the TeradataForum (TDATA-L), Manta BlueSky 
Last Modified: 30 Jun 2008