EZ Study

Actuarial Biology Chemistry Economics Calculators Confucius Engineer

SAS Tutorial in Netezza Analytics Functions
Netezza Analytics Functions

Here are the examples for various Netezza Analytics Functions:
        proc sql stimer;
        connect to netezza (server='server_name' database=database1
        user=myuser pwd=mypassword connection=global);
        execute (create table outputfile as
        select round(a.num_var,2) as num_var1, *two decimal points;
        select round(a.num_var) as num_var_int, *round to nearest integer;
        select floor(a.num_var) as num_var_less,
        *return the integer less than thar var;
        select ceil(a.num_var) as num_var_greater,
        *return the integer greater than thar var;
        select sign(a.num_var) as sign_num_var, *negative or positve;
        select cast(a.character_var as int) as numerical_var,
        /*If out of range, try bigint, numeric, or double */

        select int_var1/int_var2 as int_ratio, output only ceiling integers
        select cast(int_var1 as double)/int_var2 as int_ratio,
        /* output decimal points, instead of integers. */
        select to_number(a.character,'9999') as numerical_var,
        /*be careful , e.g. '2012M12D26' will show as '2012' for first 4 digits */
        select cast(a.numeric_var as varchar(255)) as character_var,
        select To_char(current_timestamp,'YYYY-MM-DD'),
        select To_Char( DATE '2009-12-23 23:45:58','YYYY-MM-DD'),
        select To_char(
        To_date( '2009-12-23 23:45:58','YYYY-MM-DD HH24:MI:SS'),
        'YYYY-MM-DD'), /* to_char is used for specific format */
    select cast(datetime as date) as date,cast(datetime as time) as time
        select date(current_Date) as date_today
        select (current_timestamp) as time_now
        select date_part('year', a.saledate) as year,
        select date_part('month' , a.saledate) as month,
        select date_part('day', a.saledate) as day,
        select extract(hour from a.data_timestamp) as hour,
        select extract(minute from a.data_timestamp) as minute,
        select ln(avg(value)) as log_value, not use log
        select stddev(value) as std_variance
        /* In SAS: std function returns standard deviation, */
        /* stderr=std/sqrt(n): standard error of the mean. */

        /* Get the 2nd word('1580') from the string1= '20130602-1580-2-45'
        /* scan(string1,2,'-') from SAS, = instr(string1,'-',1,2) */
substr(str1,instr(str1,'-',1,2)+1,instr(str1,'-',1,3)-instr(str1,'-',1,2)) as word2
        substr(str1,10,position('-' in substr(str1,11))) as word2
        select substring(var1,1,position('key' in var1)-1) as new_var1,
        trim(leading '0' from b.var2) as var2, c.var3
        /* remove leading zeros from the text string */

        from netezza_database1 as a
        left join netezza_database2 as b on a.keyvar1 = b.key2
        left join netezza_database3 as c on a.keyvar2 = c.key3
    /*by default, to_char() has 200 length, use substr() to truncate */
        where a.datetime >= '01/01/2010'
        where a.datetime >= '01Jan10'
        where a.datetime >= '01Jan2010 08:00:00'
        and b.var2 in ('string1', 'string2')
        group by a.var1, c.var3
        having sum(b.var2)>100
        limit 10;
        distribute on random ) by netezza; quit;

You may see the following error message if you are not joining correctly:
ERROR: CLI error trying to establish connection: [Microsoft][ODBC Driver Manager] Data source name not found and no default driver specified.
This is due to the hox-fix by SAS: http://support.sas.com/kb/20/287.html

In case if you want to use the case statement, use the following:
        execute (create table outputfile as
        select var1, case
        when var2='old1' then 'new1'
        when var2='old2' then 'new2'
        else var3
        end as new_var2

        from another_netezza_data
        where var3 is not null /* missing not working, neither for var3^=.*/
        order by var1,var3
        having new_var2=max(new_var2)
        ) by netezza; quit;

If you don't want to move those two datasets into the same database, especially if both datasets are huge, you can use the next tutorial to create a temporary file in Netezza, do the query by the temporary data.

Related links:

Continue to next: Tips for large datasets-IV: How to update temporary data in Netezza?   SAS tutorial home
Back to: Tips to handle large datasets-II: Netezza/SAS Access   Statistics tutorial home