String function in SAS

String function:

   a.        Length:

Using length function, we can find out length of the string(no. of characters include blank spaces). Length function returns numeric value.  

   b.       Index function:

Returns position of character in string. It works based on character and word wise. Note:        Specific character not available in string it returns zero. Length and index function returns numeric value.  

   c.     Scan function:

          Using scan function we can get required word from the string.  

    d.     Sub string:

Using this function, we can get part of the string from string. If requires 3 arguments.

1. Variable name
2. Starting position
3. Number of characters

e.   concatenation(combine):

it is used for combining the strings. Symbol is|| ex: data string; input pid surname $ name & : $15 age ; cards; 100  chittu Chandra  Sekhar  56 101  Kolla ram mohan rao  45 102  konda pavan kumar  34 ; Data string1; Set string; Sl = length (name); Si = index( name, ’a’); Sib= index (name, ‘ ‘); Siw = index (name, ’ra’); Sc = Scan(name,2); Sub = Substr(name,5,10); Fname=Surname|| name; Run; Proc print data =string1; Run;

Interested in mastering SAS Certification? Enroll now for FREE demo on SAS Training Program.

     f.    Compress function:

    It can be used to remove specific characters from the string. It requires 2 arguments. It is working based on the characters wise.

Syntax:

Compress =(variable name, ’character’);

Note:

   If we omit the 2^nd argument in compress function, it removes the blank spaces from the string.  

Data:

G100     col      0.005   headache and backpain G200      Aspirin  0.005   Null and skin problem G300     col      0.1   headache and Null G400      Aspirin  0.1 headache and eye problem Ex: Data one; Infile ‘d:\medi.txt’; Input group $ drug $ dos Adevent  :$12 .; Run; Data medi; Set one; If dos =0.005 then dos =5; Else dos =10; Drug1 = drug ||’-’|| dos || ‘mg’; Drug1= compress (drug1); Exad = scan(adevent ,1); Unexad = scan(adevent ,2); Drop drug dos adevent; Rename drug1 =drug; Run; Proc print data =medi; Run;  

       g.       Translate function:

It can be used to replace the required character in string. Ex: Data medi; Input pid drug$; Cards; 100 col5mg 101 col10mg 102 col15mg ; Data medi; Set medi; Drug = translate(drug.’C’.’c’); Run; Proc print data =medi; Run; Ex: Data demo; Input stno name & :$21 age color $; Cards; 100  chittu Chandra  Sekhar  23 white 101  Kolla ram mohan rao  34 black Data demo1; Set demo; Surname =scan (name,1); B_p = index (name, ‘ ’); Name= Substr(name, bp+1); Drop bp; Run; Proc print data =demo1; Run;  

Date and time function:

Ex: Data demo; Input stno scvdate :date9 Set time :time 8.; Format scvdate date 9. Scvtime time 8; Cards; 100 12jan2003  12:23:34 101  13feb2003  13:23:34 102 14feb2003  11:23:34 103  15mar2003  10:23:34 ; Data demo1; Set demo; Scv day = day(scvdate); Scv month= month(scvdate); Scv year= year(scvdate); Sc hour =hour(scvtime); Sc min =minute(scvtime); Sc sec =second (scvtime); Run; Proc print data =demo1; Run; /*To report Feb month */ Proc print data =demo1; Where scv month =2; Run;  

1. Date part function:

It can be used to get date value from the date and time variables. Syntax:     Datepart(argument)  

         2. Time part function:

    It can be used to get time value from the date and time variables. Syntax: Time part(argument) Ex: Data demo; Input stno scvdtime :datetime18; Format scvdtime datetime 18.; Cards; 100 12jan2003  12:23:34 101  13feb2003  13:23:34 ; Data demo1; Set demo; Scdate = datepart(scvdtime); Sctime = timepart(scvdtime); Format scvdate date 9. sctime time 8.; Drop scvdtime; Run; Proc print data =demo1; Run;  

   3.  Intck function:

It can be used to report difference between the date values in day intervals , month intervals or year intervals. Ex: Data medi; Input stno sdate edate; Informat sdate edate date 9; Format sdate edate date 9; Cards; 100 12jan2003  14dec2003 101  15jan2003 14oct2004 ; Data medi; Set medi; Days = inteck(‘day’, sdate edate); Run; Proc print data =medi1; Run;  

Data management:

It is one type of data reading concept using this concept, we can access data from different SAS files and arrange the data into specific order. It can be 2 types

1. Adding
2. Combine

Adding:

This concept can be done in 2 ways

1. Appending
2. Concatenation

Appending:

To add one or more data set data into a existed data set. This concept is called appending  

   Append procedure:

Using this procedure, we can do append procedure and concatenation. Syntax: Proc append base =<master> data = <trans>; Ex: Data demo 1; Input pid age gender $; Cards; 100  34  female 101  56  male ; Data demo2; Input pid age gender $; Cards; 200  56 male 201  34  female ; Data demo3; Input pid age gender $; Cards; 300  56 male 301  34  female Proc append base =demo 1 Data =demo2; Run; Proc append base =demo 1 Data =demo3; Run;  

Concatenation:

To add one or more data sets observation into another new data set this concept is called concatenation. Using append procedure we can create new data sets for concatenation. Ex: Proc append base = demo Data =demo1; Run; Proc append base = demo Data =demo2; Run; Proc append base = demo Data =demo3; Run; Case 1: If we run the append procedure multiple, observations are added in based data set. To overcome this problem use with ‘constraints’. Case 2: In appending time, additional variables are occurred in master table we can run the appending. Its default takes missing values for transition data or observation for additional variable. Case 3:     In appending time, additional variables are occurred in transition data set. We can’t run the append if we want to run we should use force option. Ex: Data demo1; Input pid age gender $; Cards; 100  34 female 101  34  male ; Data =demo2; Input pid age gender $  race $; Cards; 200  56 male  African 201  34  female Asian ; Proc append base = demo1 Data =demo2 force; Run; Case 4: If variables names are different ,then we will use rename option in appending procedure. It is renamed only for temporary purpose. Ex: Data demo1; … Data demo4; Input pid age gender $; Cards; 200  56 male 201  34  female ; Proc append base =demo 1 Data =demo4(rename =()subid=pid)); Run; Case 5: If we load transition data into master file then transition data follow the master data set or file formats. Ex: Data demo5; Input pid age gender $ vdate :ddmmyy10. Format vdate :ddmmyy10; Cards; 100  34 female   12/07/2003 101  56  male     13/08/2003 ; Data =demo6; Input pid age gender $  vdate :date 9; Format vdate :date9; Cards; 200  56 male  12sept2003 201  34  female 14oct2003 ; Proc append base = demo5 Data =demo6force; Run;  

Update statement:

It can be used to replace the master file data values with transition files based on matching variable. /* To report employee new salaries */ Ex: Data old emp1; Input eid salary; Cards; 100  4000 102  2000 103  3000 ; Data new emp2; Input eid salary; Cards; 103  6000 100  7000 103  . ; Proc sort data = old emp1; By eid; Run; Proc sort data = new emp2; By eid; Run; Data emp3; Update old emp1 new emp2; By eid; Run; After updations are completed, these updations can be stored in master file or in new file. Syntax: Data < master>/<new>; Update < master><trans>; By <match var>; Run; If we get missing value in transition data set, the corresponding value in master data set will not be changed in updation time.  

Update mode option:

Default working in update statement it avoids missing data value from the transition data or updates.  

Default value from missing check:

If we want to get missing values in updation we will use update mode = no missing check Ex: Data emp3; Update emp1 emp2 Update mode = no missing check; By eid; Run; Proc print data =emp3; Run; Case 1:      Additional variables or non – matching  variables are occurred in master file or transition file we can run the update statement. Case 2:    Additional observation or non matching observations are occurred in master file or transition file we can run the update statement. Case 3: Non- matching observation is occurred in transition file. These observation default load in master file in updation time. Ex: Data emp1; Input eid salary  if no $; Cards; 100  2000    P100 101  4000     P200 102  2000      P300 103  3000      P400; Data emp2; Input eid salary  bonus; Cards; 100  7000   300 100  9000   400 103  .            . 102   5000   700 ; Proc sort data = emp1; By eid; Run; Proc sort data = emp2; By eid; Run; Data emp3; Update emp1 emp2; By eid; Proc print data =emp3; Run; Note: Both data sets have no matching observations, in these cases if we run the update concept internally appending is running.