Processing SAS Data Sets

Transcription

1 Statistical Data Analysis 1 Processing SAS Data Sets Namhyoung Kim Dept. of Applied Statistics Gachon University nhkim@gachon.ac.kr 1

2 Using OUT Dataset OUTPUT Statement OUTPUT <OUT=SAS-data-set> <output-statisticsspecification> Create an output data set that contains specified statistics and identification variables Options OUT=: names the new output data set output-statistic-specification statistic-keyword<(variable-list)>=<names> statistic-keyword:range CV STD KURT SKEW MAX MEAN MIN MODE N and so on 2

3 Using OUT Dataset Example 8.1 3

4 Using OUT Dataset Example 8.2 4

5 Using OUT Dataset Example 8.3 NWAY option: stores observations for only the highest _TYPE_ value 5

6 Compute Ranks Using PROC RANK PROC RANK PROC RANK DATA= SAS-data-set options; BY variables; VAR variables; RANKS new-variables; RUN; Options OUT=: create an output data set TIES=: specify how to rank tied values(high, LOW, MEAN) DESCENDING: reverse the order of the rankings GROUPS=: partition observations into groups 6

7 Compute Ranks Using PROC RANK Example 8.4 7

8 Compute Ranks Using PROC RANK Example 8.5 8

9 Standardizes Variables Using PROC STANDARD Standardized values PROC STANDARD PROC STANDARD DATA= SAS-data-set options; BY variables; FREQ variable; VAR variables; WEIGHT variable; RUN; Options OUT= : specify the output data set MEANS=, STD=: specify the mean value and the standard deviation value REPLACE: replace missing values with a variable mean or MEAN= value 9

10 Standardizes Variables Using PROC STANDARD Example

11 Standardizes Variables Using PROC STANDARD Example

12 Sorting Data with PROC SORT SORT Procedure PROC SORT DATA=SAS-data-set OUT=SAS-data-set options; BY variables; RUN; There are many reasons for sorting data to organize data for a report before combining data sets before using a BY statement in another PROC or DATA step BY variables with one BY variable, SAS sorts the data based on the values of that variable with more than one variable, SAS sorts observations by the first variable, then by the second variable within categories of the first, and so on 12

13 Sorting Data with PROC SORT Example 8.8 DATA= and OUT= options specify the input and output data sets if you don t specify the DATA= option, then SAS will use the most recently created data set. if you don t specify the OUT= option, then SAS will replace the original data set with the newly sorted version 13

14 Sorting Data with PROC SORT Example 8.9 By default SAS sorts data in ascending order, from the lowest to highest. To have data sorted in the opposite order, add the keyword DESCENDING to the BY statement before each variable that should be sorted in reverse order. 14

15 Sorting Data with PROC SORT Other options NODUPKEY option (no duplicate key) to eliminate any duplicate observations that have the same values for the BY variables NODUPRECS option (no duplicate records) to eliminate any duplicate observations that have the same values for all variables. 15

16 Printing Data with PROC PRINT PROC PRINT PROC PRINT DATA= SAS-data-set options; BY variables; PAGEBY variables; SUM variables; SUMBY variables; VAR variables; RUN; The BY statement starts a new section for each new value of the BY variables and prints the values of the BY variables at the top of each section. The data must be presorted by the BY variables. The PAGEBY statement starts a new page for each new value of the PAGEBY variables or the BY variables. 16

17 Printing Data with PROC PRINT The SUM statement prints sums for the variables in the list. The SUMBY statement prints sums for the SUMBY variables or the BY variables. The VAR statement specifies which variables to print and the order. Without a VAR statement, all variables in the SAS data set are printed in the order that they occur in the data set. Options N : Print the number of observations in the data set, in BY groups, or both, and specify explanatory text to print with the number UNIFORM:Useeachvariable'sformattedwidthasitscolumn width on all pages ROUND : Round unformatted numeric values to two decimal places LABEL : Use variables' labels as column headings 17

18 Printing Data with PROC PRINT Example

19 Creating Your Own Formats Using PROC FORMAT PROC FORMAT PROC FORMAT options; VALUE <$>name <(format-options)> value-range-sets; RUN; At some time you will probably want to create your own custom formats especially if you use a lot of coded data. Using coded data is convenient for data entry and analysis but bothersome when it comes time to interpret the results. A better solution is to create user-defined formats using PROC FORMAT and print the formatted values instead of the coded values. 19

20 Creating Your Own Formats Using PROC FORMAT The name in the VALUE statement is the name of the format you are creating. If the format is for character data, the name must start with a $. The name can t be longer than 32 characters, it must not start or end with a number, and cannot contain any special characters except the underscore. In addition, the name can t be the name of an existing format. 20

21 Creating Your Own Formats Using PROC FORMAT 21

22 Creating Your Own Formats Using PROC FORMAT Creating Format Catalog Using Format Catalog OPTIONS FMTSEARCH=(mysas.prefer_format); Default: FMTSEARCH=(WORK.FORMATS SASUSER.FORMATS) 22