proc freq data=lib.dataset;
table var; *here you can mention single variable of multiple
                         
            variables seperated by space to get single     
       
            frequency;
run;

SAS is an analysis software which independent.
Sas is package (or) collection of tools which is a domain
based softwares.

SAS is a tool which is built using JAVA as the programming 
language.

i think the most common mistake in sas is putting the
semicolon (;) at the end of each statement

Most of us would forget to put semicolon(;) at the end of 
sas statements

TAB delimiter is seperate variables and observtons
it is use in infile statement as dlm='09'x.

proc import and proc export use a delimiter statement'
delimiter='09'x;

wn v r importing p\frm proc import frm xls.files
v use tab delimiters
it is used to set the spaces wn v r importing the xl sheet

SAS reds the data from different files like .txt ASCII .xls 
etc. For this use INFILE statement. While using this 
statment we have to check the data formated in the related 
file, i.e some may space, coma, tabs. So, if the file 
contains TAB is the delimiter, we have to use 
DLM=TAB / '09'X,(note: this is for horizantal TAB) to get 
the observations in a defined format. (for Vertical TAb, 
DLM='021'X.

we can find duplicated by using missover&scanover

hi, 
to display the dups in seperate  dataset please find the 
below program.

eg:
data dups;
input var1;
1
2
3
4
4
3
5
6
;
run;

/*to find the dups*/


data dups1;
set dups;
by var1;
if not(first.var1 and last.var1) then output;
run;

*now you can see the dups ;

If any one have any sas question please e.mail me to 

suresh.sasv9@gmail.com

thanks
suresh

There are two ways to display duplicate observations.

1.In data step, using first.var and last.var

2.Using proc sort with option dupout option

proc sort data=x out=y;
by v;
data m;
set y;
by v;
if first.v and last.v then delete;
run;


this is for extracting duplicate observations
from a dataset

two ways u can do ot.
1. proc sort with dupout option.
2. data step:
   data nodups dups;
   set sample;
   by x;
   if first. and last. then output nodups;
   else output dups;
   run;
proc print data=dups;
run;

label : assigns descriptive  lables to variables.
  lables allows upto 256 characters.
rename : changes the name of the variable.
  rename allows upto 64 characters only.

Yes i support with srilath By use label option we can give 
discriptive names to the particular charcter upto length of 
256 character and it is can given either in datastep or 
proc step. 

Rename is used for changing the name given inthe existing 
dataset and it should not exceed 64 charcters and i can be 
given only in the datastep.

1. Label is global and rename is local i.e., label statement
can be used either in proc or datastep where as rename
should be used only in datastep.

2.If we rename a variable, old name will be lost but if we
label a variable its short name(old name) exists along with
it's descriptive name.

Even after labeling the variable,we must use old name for 
further analysis.
Where as after renaming the variable we must use new name.

we can change the name of a variable using both rename and 
label options.
when we use label the change will be temporary i.e. only 
for printing purpose;
but, when we use rename option the change will be 
permanantly stored in the dataset.
and both the options can be used in both datastep and 
procstep.

Rename takes only 64characters length it can be used either 
datastep or proc step and this rename is remains only for 
the particular session once u terminate the session rename 
options vanishes.
Label takes 256 characters length and its is a global 
statment we can specify a brief name and this will remain 
for long time untill & unless user delete or cancell the 
variable.

rename is permanently variable name changes
label is just disply purpose

picture format writes a template for numerics
proc format ;
picture sno 
   low -  -1  =  '00.00' 
   0-9='9.999'
   10-99='99.99'
   100-999='999.9'
   ;

When you specify zero as the digit selector, any leading 
zeros in the number to be displayed are shown as
blanks. When nine is specified as the digit selector, the 
leading zeros are displayed in the output.

picture format is used to Create a template for printing 
numbers.
ex:
proc format  ;
picture nozeros  low -  -1  =  '00.00' (prefix='-')
               -1< - < 0  =  '99' (prefix='-.' mult=100)
                0 - < 1  =   '99' (prefix='.'  mult=100)
                1 - high  =  '00.00';
run;

no we cannot peroform ranking in webi6.5

The computation error is related to the misuse of context
operators in formula

Follow this Steps

1. Dont use the Not in function in the query.
2. Check the Governor Settings.
3. Check the table qualifying and weighting.
4. Check the client/server balancing.
5. Avoid using complex calculations

avoid complex report
check filter condition,use less filter
avoid unwanted calculation

TO improve the performence of the report move all the 
calculations and filters from report level to the frame 
work manager.

1. Check the Governor Settings.
2. Dont use the Not in function in the query.
3. Check the client/server balancing.
4. Avoid using complex calculations
5. Check the table qualifying and weighting.

We can use an aggregator to group by on deptno and create a 
newport to find the max(salary) and load dept no and 
salary,we'll get unique deot no and the max salary.

we can also use rank transformation by setting top, no of 
rank =1 and enable group by port deptno

Approximately 200,000 bytes of shared memory for each 
session slot at runtime.

12,000,000 bytes of memory to the session.

sorry i made a mistake before

it ocupied 4gb memori size 2gb for data cache & 2gb for 
index cache

The number of buffer blocks in a session = DTM Buffer 
Size / Buffer Block Size. Default settings create enough 
buffer blocks for 83 sources and targets. If the session 
contains more than 83, you might need to increase DTM 
Buffer Size or decrease Default Buffer Block Size.

(total number of sources + total number of targets)* 2] = 
(session buffer blocks)

(session Buffer Blocks) = (.9) * (DTM Buffer Size) / 
(Default Buffer Block Size) * (number of partitions)

DTM BUFFER SIZE :12MB
BUFFER BLOCK SIZE :64KB

Hi,

Steps:-

1.Load ur source table into source analyzer.

2.generate the target tables.

3.in the maping designer take 2 filter transformations and 
write these queries in each of the filter transformations 

for even no of records:- select * from <tablename> where 
(rowid,1) in (select rowid, mod(rownum,2) from <tablename>)

for odd no of records:- select * from <tablename> where 
(rowid,0) in (select rowid, mod(rownum,2) from <tablename>)\

4. connect the output ports of this transformation to their 
respective output tables.

Note:-
send your responses to suriaslesha_sreekar@yahoo.co.in

Can it be done in this way....

1)Drag source to mapping designer
2)Take a router transformation .consider EMP table in which
i am using EMPNO. In group1, assign the condition as
mod(empno,2)=0 which gives even numbers and in group2,
assign the condition as mod(empno,2)!=0 which gives odd numbers.
3)connect group1 to one target and group2 to another target.

If i am wrong please tell me....

We can do this in the following.

Take the sequence generator t/r
set the properties like
Start value 1
End value 2
and also Enable cycle option

connect the nextval port to the router T/r 
port "SGTNO"(created by you) and also connect the ports 
from source qualifier to the router t/r

Now given the first condition like
SGTNO=1 then go to first target
otherwise go to second target(here no need to mention the 
second condition).

If I went wrong, please let me know

Thanks
Anand Kumar

1. drag source and targets in to mapping designer work space
2. from t/r devoloper take sequence genarato t/r,exp t/r 
and router t/r.
3. in seqg t/r give startvalue 1 ,increment by 1
   give netval to a newpor in exp t/r
4. drag all ports of sq to exp in addition to newport
5. in router t/r create one group name as odd 
   give condetion mode(newport/2)!=0
6. give from group odd to t1
   and defoult to t2.

An operational data store (ODS) is a type of database often 
used as an interim area for a data warehouse. Unlike a data 
warehouse, which contains static data, the contents of the 
ODS are updated through the course of business operations. 
An ODS is designed to quickly perform relatively simple 
queries on small amounts of data (such as finding the 
status of a customer order), rather than the complex 
queries on large amounts of data typical of the data 
warehouse. An ODS is similar to your short term memory in 
that it stores only very recent information; in comparison, 
the data warehouse is more like long term memory in that it 
stores relatively permanent information.

But in staging we are storing current as well as historic 
data. This data might be a raw and then need cleansing and 
transform  before load into datawarehouse.