Welcome to Top 10 SAS Functions

Size: px

Start display at page:

Download "Welcome to Top 10 SAS Functions"

Charlene Carson
5 years ago
Views:

1 Welcome to Top 10 SAS Functions

2 Goal and Agenda By the end of this meeting, you will understand 10 key SAS functions purpose, value and features. What are SAS functions? Why use them? Use Case Manipulating character data Working with dates (derive durations) Summarizing and rounding numeric data

3 SAS functions conduct a wealth of data manipulation operations so you can transform raw data into intelligence and insight. Categories of functions Character Currency Financial Trigonometric Descriptive statistics Date/time/temporal What to remember Why? Prepare data for reporting and analytics enhance / combine / convert data for reporting and analytics

4 What are SAS Functions? Why use SAS Functions? What? SAS functions perform computations, data manipulation, and enhancement Character Numeric Dates/temporal Mathematics Why? You need to transform raw/operational data into report- and analytics-ready structures DATA step programming SQL

5 Why these 10? What are the top 10 functions? 2017 surveys of SAS technical support (what customers submit most often) a 2016 survey on communities.sas.com Other trusted sources Which 10? character manipulation (SUBSTR, SCAN, CATX, FIND) temporal / date / time (DATEPART, DATEDIF) data conversion (PUT, INPUT) numeric (SUM, ROUND)

6 Where to find SAS Functions 1 of 2

7 Where to find SAS Functions 2 of 2

8 Use Case Raw data before functions The human resources team at a fictitious organization needs to know more about its divisions employees. employee attrition tenure total compensation

9 Use Case intelligence after using functions 1 of 2 Without formats

10 Use Case intelligence after using functions 2 of 2 With formats as shown in the PROC PRINT proc print; format newterm newhire date9. totalcomp dollar15.; run;

11 About the use case data - 1 of 2 IDNO is stored as a numeric value. The last digit of IDNO represents a division within the company. HIREDATE is a SAS datetime value. TERMDATE is a character value.

12 About the use case data - 2 of 2 Here are the attributes of the data:

13 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

14 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division a) Since IDNO is stored as numeric, it must be converted to character via the PUT function. b) Once IDNO becomes character data, its last character can be extracted via the SUBSTR function.

15 PUT (source, format) PUT Syntax converts the expression to a character string - always returns a character value source argument can be character or numeric. format contains the SAS format that you want applied to the variable or constant that is specified in the source. To override the default alignment, you can add an alignment specification to a format: L left aligns the value. C centers the value. R right aligns the value.

16 Convert Numeric to Character PUT function 1. Extract the last character of IDNO in order to identify the division a) Since IDNO is stored as numeric, it must be converted to character via the PUT function. idnoc = put(idno, 4.);

17 Extract from a Character String SUBSTR function 1. Extract the last character of IDNO in order to identify the division b) Once IDNO becomes character data, its last character can be extracted via the SUBSTR function.

18 SUBSTR Syntax Extracts a substring from an argument - returns the characters from start to end SUBSTR (string, position<, length>) string specifies any SAS character expression. position specifies a numeric expression that is the beginning character position. length specifies a numeric expression that is the length of the substring to extract. If you omit length, SAS extracts the remainder of the expression.

19 Extract from a Character String divid = substr (idnoc,4,1);

20 Excerpted DATA Step Code Step 1-Extract Division Number and Create Division Name data newinfo; set info; idnoc=put(idno, 4.); divid=substr(idnoc,4,1); length divname $ 25; if divid='1' then divname='best Gadget Toys'; else if divid='2' then divname='best Gadget Tools'; else if divid='3' then divname='best Gadget Appliances'; run;

21 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

22 How to transform the raw data into intelligence? 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable). a) Use the SCAN function to separate the values into and LASTNAME and FIRSTNAME. b) Connect LASTNAME and FIRSTNAME back together in the form FIRSTNAME LASTNAME using the CATX function.

23 SCAN Syntax Selects a given word from a character expression SCAN (string,n<, delimiter(s)>) string specifies any character expression. n specifies a numeric expression that produces the number of the word in the character string you want SCAN to select. delimiter specifies a character expression that produces characters that you want SCAN to use as a word separator in the character string.

24 Extract and separate last name and first name SCAN function 2a) Use the SCAN function to separate the values into LASTNAME and FIRSTNAME. lname = scan(name,1,', '); fname = scan(name,2,', ');

25 CATX Syntax Concatenates character strings, removes leading and trailing blanks, and inserts separators CATX (separator, string-1 <,...string-n>) separator specifies a character string that is used as a separator between concatenated strings. string specifies a SAS character string. The CATX function returns a value to a variable, or returns a value in a temporary buffer.

26 Re-arrange as first name last name CATX function 2b) Connect LASTNAME and FIRSTNAME back together in the form FIRSTNAME LASTNAME using the CATX function.. newname = catx(' ',fname,lname);

27 data newinfo; set info; idnoc=put(idno, 4.); divid=substr(idnoc,4,1); Excerpted DATA Step Code Step 2 - Create New Name length divname $ 25; if divid='1' then divname='best Gadget Toys'; else if divid='2' then divname='best Gadget Tools'; else if divid='3' then divname='best Gadget Appliances'; lname=scan(name,1,', '); fname=scan(name,2,', '); newname=catx(' ',fname,lname); run;

28 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

29 How to transform the raw data into intelligence? 3. Search for misspelled name Use the FIND function to identify a misspelled name and write a message to the SAS log. Make note of the misspelling in the SAS log

30 FIND Searches for a specific substring of characters within a character string that you specify FIND (string,substring<,modifiers>) string specifies a character constant, variable, or expression that will be searched for substrings. substring is a character constant, variable, or expression that specifies the substring of characters to search for in string. Modifiers is a character constant, variable, or expression that specifies one or more modifiers. The following modifiers can be in uppercase or lowercase: I ignores character case during the search. If this modifier is not specified, FIND only searches for character substrings with the same case as the characters in substring. Tip: Enclose a literal string of characters in quotation marks.

31 Search for misspelled name FIND function if find(newname, 'Lau', 'i') > 0 then put 'value found ' newname; Partial SAS Log

32 data newinfo; set info; idnoc=put(idno, 4.); divid=substr(idnoc,4,1); Excerpted DATA Step Code Step 3 - Find Misspelled Name length divname $ 25; if divid='1' then divname='best Gadget Toys'; else if divid='2' then divname='best Gadget Tools'; else if divid='3' then divname='best Gadget Appliances'; lname=scan(name,1,', '); fname=scan(name,2,', '); newname=catx(' ',fname,lname); if find(newname, 'Lau', 'i')>0 then put 'value found ' newname; run;

33 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

the SAS date portion of HIREDATE using the DATEPART function b) Convert TERMDATE to

34 How to transform the raw data into intelligence? 4) Determine the number of days (years) between HIREDATE and TERMDATE a) Extract the SAS date portion of HIREDATE using the DATEPART function b) Convert TERMDATE to a SAS date using the INPUT function c) Calculate the number of days (years) between the 2 new hire and termination dates

35 What are SAS date, time and datetime values? SAS stores date values as numeric SAS date value a value that represents the number of days between January 1, 1960, and a specified date. SAS time value a value representing the number of seconds since midnight of the current day. SAS time values are between 0 and SAS datetime value a value representing the number of seconds between January 1, 1960 and an hour/minute/second within a specified date.

36 SAS Date Values SAS stores date values as numeric 01JAN JAN JAN1961 store display 01/01/ /01/ /01/1961

37 SAS Date Values 01Jan1960 Valid from 1582 A.D. to A.D. Many formats available for grouping/reporting 0 Date. 12OCT17 Date9. 12OCT2017 Worddate. October 12, 2017 Weekdate. Thursday, October 12, 2017 Month. 10 Monname. October Monname3. Oct monyy7. OCT2017

38 SAS Datetime Values SAS stores datetime values as numeric values 01JAN59:00:00:00 01JAN60:00:00:00 01JAN61:00:00:00 store display 01JAN59:12:00:00 AM 01JAN60:12:00:00 AM 01JAN61:12:00:00 AM A SAS datetime value is the number of seconds between January 1, 1960 and an hour/minute/second within a specified date.

39 Extracts the date from a SAS datetime value DATEPART DATEPART (datetime) datetime specifies a SAS expression that represents a SAS datetime value.

40 Extract the date portion of hire date DATEPART function 4a) Extract the SAS date portion of HIREDATE using the DATEPART function newhire = datepart(hiredate); New SAS Date Value New SAS Date Value with Format

41 Convert termination date portion to a SAS date INPUT function 4b) Convert TERMDATE to a SAS date using the INPUT function New SAS Date Value New SAS Date Value with Format

42 INPUT Syntax INPUT (character-expression, informat) Converts a string expression using the specified informat Often used to convert character to numeric

43 Convert termination date portion to a SAS date INPUT function 4b) Convert TERMDATE to a SAS date using the INPUT function newterm = input(termdate,date9.); New SAS Date Value New SAS Date Value with Format

44 Calculate the number of days between the hire and termination dates DATDIF function 4c) Calculate the number of days (years) between the 2 new hire and termination dates minus

45 DATDIF Syntax Returns the number of days between two dates DATDIF (sdate,edate,basis) sdate specifies a SAS date value that identifies the starting date. edate specifies a SAS date value that identifies the ending date. basis identifies a character constant or variable that describes how SAS calculates the date difference. 30/360 or 360 specifies a 30 day month and a 360 day year. ACT/ACT or Actual uses the actual number of days between dates.

46 Calculate the number of days (years) between the 2 new hire and termination dates DATDIF Now that both dates are in the form of a SAS date value, we can find the difference between the 2 dates using the DATEDIF function: if newterm ne. then do; dayofserv=datdif(newhire,newterm, 'ACT/ACT'); yearofserv= yrdif(newhire, newterm, 'ACT/ACT'); end; DATDIF

47 data newinfo;... Excerpted DATA Step Code Step 4 Derive Years (Days) of Service newterm=input(termdate,date9.); newhire=datepart(hiredate); if newterm ne. then do; dayofserv=datdif(newhire,newterm, 'ACT/ACT'); yearofserv= yrdif(newhire, newterm, 'ACT/ACT'); end; run;

48 How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division. 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

together. a) The SUM function will add up the nonmissing values.

49 How to transform the raw data into intelligence? 5. Calculate the total compensation by adding SALARY, BONUS and MERIT together. a) The SUM function will add up the nonmissing values. b) The ROUND function will express the total compensation as a whole number without decimal positions.

50 SUM Syntax SUM (argument1, argument2,...) Argument(s) are numeric. The argument list can consist of a variable list, which can be preceded by OF. If all the arguments have missing values, the result is a missing value. x1=sum (4,9,3,8); 24 x2=sum (4,9,3,8,.); 24

51 Calculate Total Compensation SUM Function 5a) Calculate the total compensation by adding SALARY, BONUS and MERIT together. totalcomp = sum(salary, bonus, merit);

52 Calculate Total Compensation ROUND Function 5b) The ROUND function will express the total compensation as a whole number without decimal positions.

53 ROUND Syntax Rounds the first argument to the nearest multiple of the second argument, or to the nearest integer when the second argument is omitted ROUND (argument <,rounding-unit>) argument is a numeric constant, variable, or expression to be rounded. rounding-unit is a positive, numeric constant, variable, or expression that specifies the rounding unit.

54 Calculate Total Compensation ROUND Function 5b) The ROUND function will express the total compensation as a whole number without decimal positions. totalcomp = round(totalcomp);

55 data newinfo;... newterm=input(termdate,date9.); newhire=datepart(hiredate); Excerpted DATA Step Code Step 5 - Calculate Total Compensation if newterm ne. then do; dayofserv=datdif(newhire,newterm, 'ACT/ACT'); yearofserv= yrdif(newhire, newterm, 'ACT/ACT'); end; totalcomp=sum(salary, bonus, merit); totalcomp=round(totalcomp); run;

56 Review Raw data before functions The human resources team at a fictitious organization needs to know more about its divisions employees. employee attrition tenure total compensation

57 Review intelligence after using functions 1 of 2 Without formats

58 Review intelligence after using functions 2 of 2 With formats as shown in the PROC PRINT proc print; format newterm newhire date9. totalcomp dollar15.; run;

59 Review - How to transform the raw data into intelligence? 1. Extract the last character of IDNO in order to identify the division 2. Separate the name values into LASTNAME and FIRSTNAME, then combine them together as FIRSTNAME LASTNAME (one new variable) 3. Search for misspelled name 4. Determine the number of days (years) between HIREDATE and TERMDATE 5. Calculate total compensation

60 Final Data Step Code 1 of 2 data newinfo(drop=idno name lname fname salary bonus merit hiredate termdate); set info; idnoc=put(idno, 4.); divid=substr(idnoc,4,1); length divname $ 25; if divid='1' then divname='best Gadget Toys'; else if divid='2' then divname='best Gadget Tools'; else if divid='3' then divname='best Gadget Appliances'; lname=scan(name,1,', '); fname=scan(name,2,', '); newname=catx(' ',fname,lname); if find(newname, 'Lau', 'i')>0 then put 'value found ' newname;

61 newterm=input(termdate,date9.); newhire=datepart(hiredate); Final Data Step Code 2 of 2 if newterm ne. then do; dayofserv=datdif(newhire,newterm, 'ACT/ACT'); yearofserv= yrdif(newhire, newterm, 'ACT/ACT'); end; totalcomp=sum(salary, bonus, merit); totalcomp=round(totalcomp); run;

62 SAS functions conduct a wealth of data manipulation operations so you can transform raw data into intelligence and insight. Categories of functions Character Currency Financial Trigonometric Descriptive statistics Date/time/temporal What to remember Why? Prepare data for reporting and analytics enhance / combine / convert data for reporting and analytics

63 Hungry for more? Recommended Resources Papers Training SAS Programming 2: Data Manipulation Techniques Connect Community Blog Books and Documentation x0an0z2dzozh2ouzm.htm

Top 10 SAS Functions in A brief summary of SAS Communities Survey - by Flora Fang Liu

Top 10 SAS Functions in A brief summary of SAS Communities Survey - by Flora Fang Liu Top 10 SAS Functions in 2017 A brief summary of SAS Communities Survey - by Flora Fang Liu 1 What are SAS Functions? Why use SAS Functions? What? SAS functions perform computations, data manipulation,