Sans titre - WeAreCAS

Data Analysis

Type : CREATION_INTERNE

Examples use generated data (datalines) to ensure code autonomy.

1 Code Block

DATA STEP / PROC PRINT Data

Explanation :
This example illustrates a one-to-one merge of two data sets, 'animal' and 'plantG', which have an equal number of observations. The MERGE statement is used without a BY statement. The values of the 'common' variable in 'plantG' (the last specified data set) overwrite those from 'animal' at the sixth observation, where they differ. Non-common variables are added, and all observations are included in the result.

Copied!

1	DATA animal;
2	INPUT common $ animal $;
3	DATALINES;
4	a Ant
5	b Bird
6	c Cat
7	d Dog
8	e Eagle
9	f Frog
10	;
11	DATA plantG;
12	INPUT common $ plant $;
13	DATALINES;
14	a Apple
15	b Banana
16	c Coconut
17	d Dewberry
18	e Eggplant
19	g Fig
20	;
21	DATA merged;
22	MERGE animal plantG;
23	RUN;
24
25	PROC PRINT DATA=merged; RUN;

2 Code Block

DATA STEP / PROC PRINT Data

Explanation :
This scenario shows a one-to-one merge between 'animal' (6 observations) and 'plantMissing' (3 observations). The MERGE statement without a BY statement will process all records from each data set by combining them line by line. Missing observations in the shorter data set ('plantMissing') result in missing values in the corresponding columns of the resulting data set where there is no implicit match, as the merge continues until all observations from all data sets are processed. If the 'SET' statement had been used, the program would have stopped after reading the last record of the smaller data set.

Copied!

1	DATA animal;
2	INPUT common $ animal $;
3	DATALINES;
4	a Ant
5	b Bird
6	c Cat
7	d Dog
8	e Eagle
9	f Frog
10	;
11
12	DATA plantMissing;
13	INPUT common $ plant $;
14	DATALINES;
15	a Apple
16	b Banana
17	c Coconut
18	;
19
20	DATA merged;
21	MERGE animal plantmissing;
22	RUN;
23	PROC PRINT DATA=merged; RUN;

3 Code Block

DATA STEP / PROC PRINT Data

Explanation :
This example illustrates the potentially undesirable results of a one-to-one merge without a BY statement when data sets ('animalDupes', 'plantDupes') contain duplicate values for the 'common' variable. One-to-one merges are designed for data with a one-to-one relationship. In this scenario (one-to-many or many-to-one relationships), the merge occurs line by line, overwriting common variable values and potentially leading to loss of information or incorrect combination of observations.

Copied!

1	DATA animalDupes;
2	INPUT common $ animal $;
3	DATALINES;
4	a Ant
5	a Ape
6	b Bird
7	c Cat
8	d Dog
9	e Eagle
10	;
11
12	DATA plantDupes;
13	INPUT common $ plant $;
14	DATALINES;
15	a Apple
16	b Banana
17	c Coconut
18	c Celery
19	d Dewberry
20	e Eggplant
21	;
22
23	DATA merged;
24	MERGE animalDupes plantDupes;
25	RUN;
26	PROC PRINT DATA=merged; RUN;

4 Code Block

DATA STEP / PROC PRINT Data

Explanation :
This example demonstrates the undesirable results of a one-to-one merge without a BY statement when data sets ('animalMissing', 'plantMissing2') have different values for their common variable 'common'. The merge occurs line by line, and values from 'plantMissing2' overwrite those from 'animalMissing' for the 'common' variable. Without explicit value matching via a BY statement, the data combination can be incorrect and may not reflect the expected relationships between observations.

Copied!

1	DATA animalMissing;
2	INPUT common $ animal $;
3	DATALINES;
4	a Ant
5	c Cat
6	d Dog
7	e Eagle
8	;
9	DATA plantMissing2;
10	INPUT common $ plant $;
11	DATALINES;
12	a Apple
13	b Banana
14	c Coconut
15	e Eggplant
16	f Fig
17	;
18	DATA merged;
19	MERGE animalMissing plantMissing2;
20	RUN;
21	PROC PRINT DATA=merged; RUN;

This material is provided "as is" by We Are Cas. There are no warranties, expressed or implied, as to merchantability or fitness for a particular purpose regarding the materials or code contained herein. We Are Cas is not responsible for errors in this material as it now exists or will exist, nor does We Are Cas provide technical support for it.

Retour à la liste