Different types of input methods in SAS

Different types of input methods: 

1. List input method (or) free formatted input
2. Named input method
3. Column input method
4. Format input method
5. Absolute input method
6. Mixed input method

 

List input method:

 data values are represented by at least one blank space. Any character data value does not have embedded space. In this cases we use list input method.

storage capability for list input method:

         It stores 8 characters for character variable and 12 numbers for number variable.  

Internal storage:

1 char – 2 bytes 1 number – 1 byte Ex: Data demo; Infile ‘d:\ one.txt’; Input pid name $ age color $; Run; Proc print data =demo; Run; 100 kumar 78 white 101 kiran 89 black 102 lava 78 white INFILE options:

1. Dlm (delimiter)
2. Dsd (data separated delimiter, data specified delimiter)
3. Flow over
4. Stop over
5. Miss over
6. Turn over
7. Firstobs
8. Obs
9. Scan over

Using infile option we can read the data in proper order. All infile options supports list input method.

 

Inclined to build a profession as SAS Developer? Then here is the blog post on SAS Certification Program.

Dlm option:

It is used to indicate delimiters in raw data. Ex: Data demo 5; Infile cards dlm = ‘, & $’ ; Input pid name $ age color $; Cards; 100,  kumar rao,78,white 101 ,kiran ,89 $black 102 ,lava, 78 & white Proc print data =demo5; Run; Note:    If we use the dlm options we should indicate all the delimiter including default delimiter(space). Dsd:

• data separated delimiter:

                   In raw data, data values are separated by comma we will use dsd instead of dlm.

• Data sensitive (or) data specified delimiter:

In raw data, character data value available with quotation for manual checking, to read this raw data into SAS file without quotation with data sensitive delimiter. Ex: Data demo 6; Infile cards dsd ; Input pid name $ age color $; Cards; 100,  kumar rao,78,white 101 ,kiran ,89 “black” 102 ,lava, 78 & white

3. kumar,67,”black”

; Proc print data =demo6; Run; Ex: Data demo 6; Infile cards dsd  dlm = ‘$’; Input pid name $ age color $; Cards; 100,$ kumar rao$78,white 101 ,kiran ,89 “black” 102 $lava, 78 & white

3. kumar,67,”black”

; Proc print data =demo6; Run;  

Low over:

Default nature of the SAS is controlled by SAS system to read one line for one observation. If any value is missing in raw data it tried to read next data value. Some cases it is success and some cases it is failure. table align="left" border="1" cellpadding="0" cellspacing="0">

Missing	Next	Rating
Ch	N
N	Ch	*
Ch	Ch
N	N

            Character or numeric value is missing in raw data then character missing value is black and numeric missing value is dot. Ex: Data sasuser.demo; Infile cards; Input pid age color $ race $ Cards; 100  23  white  Asian 101  .      white Asian 102  78    . African ; Proc print data =demo; Run;

 

Stop over:

When error is occur in raw data in reading stop over nature tries to stops the reading. Errors are 2 types

1. Compilation (or) syntax error
2. Execution (or) data error

 

Syntax error:

We can identify in compilation

• Forget the semicolon at the end of the SAS statement
• Unfortunately, writing data set block in procedure block, procedure block in data set block.

 

Data error:

This can be identified by in execution. This is execution error (or) data error

• Variables request to numeric value but you will send character value.
• Row or observation indicates 4 data values but you mention the input statement 5 data values are observed.

Miss over: In raw data, at the end of the values are missing. No need to use dot(.). we should use miss over option. Miss over is controlled by flow over. Ex: Data demo; Infile cards miss over; Input pid age color $ race $ Cards; 100  23  white 101  63  white Asian 102  78 103 23 black African 104 45 black 105 56 white Asian 106 ; Proc print data =demo; Run; Syntax: /* comment */ *comment ; /*use for multiple infile options*/ Data demo3; Infile cards Dsd miss over dlm=’, $’; Input pid age color $ race $ Cards; 100 , 23 $  white 101  ,67 $  white,” Asian” 102 , 78 103 $23, black “African” ; Proc print data =demo; Run;

 

Turn over:

It works like a miss over. It adjust vary of the length for required variable and have the displace and reduce the storage place.  

Firstobs:

     We can read part of the data in sequential order, firstobs is based on lines. Data demo; Infile cards firstobs =2 obs=4; Dsd miss over dlm=’, $’; Input pid age color $ race $ Cards; 100  23  white  Asian 101  63  white Asian 102  78    white Asian 103 23 black African 104 45 black African 105 56 white Asian ; Proc print data =demo; Run;  

Scan over:

It is used to read required data values based on key data values. It is used only for character not for numeric data. Ex: Data clinical; Infile cards scan over; Input @ ‘appolo’ trail $ year sub; Cards; Appolo phase 1 1996 28 Nims phase 1 1997 30 Appolo phase 2 1997 290 Nims phase2 1998 250 ; Proc print data = appolo; run;  

To read tab files data:

Tabspace delimiter indicated by ‘09’ x Data clinical; Infile ‘d:\tab file.txt’ dlm =’09’x; Input center $ trail $ year sub; run; Proc print data = clinical; run;   Note: If we run the data set application without data set name, it default takes default name data 1,..data n Space between the SAS words is called token   Note:

1. Unfortunately space is occurred in between data set name. in this SAS recognize multiple data sset name.
2. Using data set name you can create multiple data set names.

1. Data demo graphic

Input name $ age; Cards; Uma 45 Kiran 78 ;

2. Data sasuser.demographic;

Input name $age; Cards; Uma 45 Kiran 78 ;  

Input methods:

& modifier list input method : modifier list input method ~ (tilde) modifier list input method  

& modifier list input method :

It is used to indicate 2 more blank spaces, delimiter between the required data values.  

: modifier list input method:

          It is used to increase the storage capacity with out changing the order if data. Data demo 6; Input pid name & $ : 13 age color $; Cards; 100 kiran  kumar 89 white 101 pavan 90 white 102 lava kumar 89  white ; Proc print data =demo6; run;  

~ modifier list input method:

Tilde modifier can be used to avoid dsd option for required variables. Ex: Data medi; In file cards dsd; Input pid drug $ adevents : ~ $ 1+; Cards; 100,”col 5 mg”, “fever,cold” 101,”col 5 mg”, “cold, headache” ; Proc print data = medi; run;  

To read non – standard data:

Ex: Date time amounts (numerical data)  

Informat technique:

            It is used in data set block. Using informat technique , we can read non standard data into standard data. Using informat technique we can read data values into number value(standard format). This number otherwise called SAS date value. Ex: Data medi; Input pid Jdate; Informat Jdate ddmmyy10; Cards; 200  01/01/1960 201  01/01/1961 202  12/12/1970 203 23/03/1989 ; Proc print data =medi; run; SAs date value – no.of days difference between sas discover date to your loading date or presenting date.    

Format technique:

It can be used to convert standard data into non – standard data for reporting. Format statement can be written in the procedure block. 01/01/1960 – informat -> standard (number) ->0 0 -> format -> non – standard format – 01/01/1960-> (reporting) Ex: Data medi; Input pid Jdate; Informat Jdate ddmmyy10; Cards; 200  01/01/1960 201  01/01/1961 202  12/12/1970 203 23/03/1989 ; Proc print data =medi; Format jdate ddmmyy10; run; Syntax: Informat<variable name><informat tech> Format <variable name><format tech>

• Date values informat                    format

30/02/2003                                                  ddmmyy10. Or s10 30-02-2003                                                          ddmmyyD10. 30:02:2003                                                          ddmmyyc10. 30.02.2003                    ddmmyy10             ddmmyyp10. 30022003                                                             ddmmyyB10 Ex: Data medi; Input pid Jdate; Informat Jdate ddmmyy10; Cards; 200  01/01/1960     12.01-1961 201  01/01/1961       13:04:1961 202  12/12/1970       14:02:1971 203 23/03/1989           2511190 ; Proc print data =medi; Format jdate ddmmyyD10; Ldate ddmmyyC10; run; we can report required variables in required format we can’t report each and every data value in required format.

• Date values informat                    format

30/02/03                                                      ddmmyy10S8. 30-02-03                                                               ddmmyyd8. 30:02:03                                                               ddmmyyC8. 30.02.03                    ddmmyy10                 ddmmyyP8. 3002203                                                                ddmmyyB8.

• Date value informat                  format

02/30/2003                   mmddyy10. ddyymm10 yyddmm10       wrong application mm/yy/dd yy/mm/dd Ex: Data medi; Input pid drug $ sdate edate; Informat sdate mmddyy10. Edate mmddyy 8.; Cards; 200   5mg  03/23/2003     04/25/03 201   10mg  02/25/2003     05/26/03 ; Proc print data = medi; Format sdate ddmmyD10. Edate mmddyyC10.; Run; 4)

Date value	informat	Format
23oct2003	Date9	Date9
23 oct03	Date7	Date7
Dec2003	Monyy7	Monyy7
Dec 03	Monyy5	Monyy5

Ex: Data medi; Input pid drug $ sdate edate adsdate; Informat sdate  date9 Edate date 7 Adsdate  monyy7.; Cards; 100   5mg  12oct2003     13dec03  jan2004 101   10mg  13 jun2003     26dec03   feb2004 ; Proc print data = medi; Format sdate date9. Edatedate Adsdatemonyy7.; run; report available in paper filr or electronic files. Another date value(Julian date value(or)code level date value)          

Julian date:

It is also called code level date. In this code level date the 1^st 4 digits indicate years, next 3 digits indicated no.of days completed in that year. Normal year 365 days, leap year 366 days. Max or min value for this is 7. Ex: Data adevent; Input pid ad $ sdate; Informat sdate  Julian7.; Cards; 100   headache   2003032 101   cold              2003145 102    skinprb        2003178 ; Proc print data = adevent; Format sdate date9. run; only formats – reporting purpose word date 18(max), week date 24(min) word date 15(min), week date 30(max) ex: Data adevent; Input pid ad $ sdate; Informat sdate  Julian7.; Cards; 100   headache   2003032 101   cold              2003145 102    skinprb        2003178 ; Proc print :adevent; Format sdate weekdate 24; Run;   Amounts: represented by this application

Date values	informat	Format
2,225,000	Comma9	Comma9
$2,225,000	Dollar10	Dollar10

Ex: Data emp; Input eid salary pf; Information salary comma 6 Pf dollar 10.; Cards; 100   23,000         $1,345,000 101   34,000          $1,234,678 ; Proc print data =emp; Format salary comma 6. Pf dollar 10.; Run;   Words w: Using with words we can report amounts in words w   Note:   Based on client requirement we should create informat and format using format procedure.   Times:

Date values	Informat	format
12hr -10:12:23AM02:23:52 PM	Time10	Time ampm12(etime-stime)
29hr- 22:10:2010:12:20	Time 8	Time 8
12 oct 2003:02:23:30pm12oct2003:15:23:30	Datetime20Datetime18	Dateampm22Datetime18

  Ex: Data trt; Input pid stime etime; Informat stime time10. Etime time8.; Cards; 200  10:23:34 am   15:34:19 201    02:23:34pm   18:23:34 ; Proc print data =trt; Format Stime time ampm12. Etime  time8.; Run;   Note: In raw data, only time values are available. in these cases SAS default recognizes data values is SAS discovered date. Ex: Data trt; Input pid stime etime; Informat stime datetime20. Etime datetime18.; Cards; 200  12oct2003:10:23:34am       15dec2003:15;34:19 201   13dec2003:02:23:34pm       30dec2003:18:23:94 ; Proc print data =trt; Format Stime  datetime ampm12. Etime  datetime18.; Run;  

To read percentage data:

Ex: Data bank; Input loantype $irate; Informat irate percent3.; Cards; Personal 5% House 20% Education  2% ; Proc print data =bank; Format irate percent5.; Run;   We can run old technology in new technology, but new technology can n’t run in old technology.   Note: If informat technique written in input statement then we should use colon modifier. Ex: Data bank; Input loantype ; Irate: percent3.; Cards; Personal 5% House 20% Education  2% ; Format can be assigned in 3 ways 1.Temporary 2.parmanent If we write formats in procedure block these formats are called temporary If we write formats in data set block. These formats are called permanent. Permanent formats can be changed for “reporting’. Ex: Data sasuser. Medicine; Input pid sdate : date 9. Edate : ddmmyy 10.; Format sdate date9 Edate:ddmmyy10; Cards; 100 12jan2003   13-10-1003 101  14nov2003   15-12-2003 ; Proc print data =sasuser.medicine; Run;    

SAS data set outside of the SAS environment:

Ex: Data ‘d:\demo’; Input pid age; Cards; 100 89 234 34 ; Proc print data =’d:\demo’; Run;

To save the SAS application:

SAS application can be stores or saved inside or outside of SAS environment.

 

Outside of the SAS environment:(use with menu drive)

If we SAS application outside(p.c) it default takes extension name as.sas(open program)  

Inside of the SAS:

If we want to save SAS application inside of the SAS environment, 1^st of all we can create catalogue in required library. Save your application it default takes file name as source. For indepth knowledge on SAS, click on below

• Normalization in SAS
• SQL Aggregations & DataSet(Row & colummn operations) in SAS
• SQL Statements in SAS
• SAS Interact With Oracle
• SAS Interview Questions & Answers