Factor Analysis Example (PROC FACTOR)

The script begins by creating a dataset named `SocioEconomics` containing indicators for different localities (Population, School, etc.). Then, it executes the `PROC FACTOR` procedure three times to perform a factor analysis. Each execution uses the maximum likelihood method (`method=ml`) but varies the number of factors to extract (1, 2, then 3), allowing for model comparison. The `heywood` option is used to handle cases where communalities exceed 1.

Data Analysis

Type : CREATION_INTERNE

The `SocioEconomics` data is created directly in the code via a DATA step with a `datalines` statement. It does not come from SASHELP or an external source.

1 Code Block

DATA STEP Data

Explanation :
This block creates the `SocioEconomics` table in memory. The data is included directly in the script via the `datalines` statement, a common method for small example datasets.

Copied!

1	DATA SocioEconomics;
2	INPUT Population School Employment Services HouseValue;
3	DATALINES;
4	5700 12.8 2500 270 25000
5	1000 10.9 600 10 10000
6	3400 8.8 1000 10 9000
7	3800 13.6 1700 140 25000
8	4000 12.8 1600 140 25000
9	8200 8.3 2600 60 12000
10	1200 11.4 400 10 16000
11	9100 11.5 3300 60 14000
12	9900 12.5 3400 180 18000
13	9600 13.7 3600 390 25000
14	9600 9.6 3300 80 12000
15	9400 11.4 4000 100 13000
16	;

2 Code Block

PROC FACTOR

Explanation :
This procedure performs a first factor analysis on the `SocioEconomics` table. `method=ml` specifies the use of the maximum likelihood method. `n=1` constrains the model to extract only one factor. `heywood` allows the process to continue even if a communality is greater than 1.

Copied!

1	title3 'Maximum Likelihood Factor Analysis with One Factor';
2	PROC FACTOR
3	DATA=SocioEconomics method=ml heywood n=1;
4	RUN;
5

3 Code Block

PROC FACTOR

Explanation :
Similar to the previous one, this factor analysis tests a two-factor model (`n=2`) to evaluate whether a more complex structure better explains the data variance.

Copied!

1	title3 'Maximum Likelihood Factor Analysis with Two Factors';
2	PROC FACTOR
3	DATA=SocioEconomics method=ml heywood n=2;
4	RUN;
5

4 Code Block

PROC FACTOR

Explanation :
This last analysis extends the model to three factors (`n=3`), completing the comparison of factorial models to determine the most relevant underlying structure.

Copied!

1	title3 'Maximum Likelihood Factor Analysis with Three Factors';
2	PROC FACTOR
3	DATA=SocioEconomics method=ml heywood n=3;
4	RUN;
5

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