Sans titre - WeAreCAS

Data Analysis

Type : CREATION_INTERNE

Data is manually created via the DATALINES statement in the first DATA step.

1 Code Block

DATA STEP Data

Explanation :
Creation of the 'cash' dataset containing two variables: School (group) and Money (numeric value).

Copied!

1	DATA cash;
2	INPUT School Money;
3
4	DATALINES;
5	0 34
6	0 1200
7	0 23
8	0 50
9	0 60
10	0 50
11	0 0
12	0 0
13	0 30
14	0 89
15	0 0
16	0 300
17	0 400
18	0 20
19	0 10
20	0 0
21	1 20
22	1 10
23	1 5
24	1 0
25	1 30
26	1 50
27	1 0
28	1 100
29	1 110
30	1 0
31	1 40
32	1 10
33	1 3
34	1 0
35	;

2 Code Block

PROC TTEST

Explanation :
Execution of the initial Student's t-test on observed data to calculate the actual mean difference between groups.

Copied!

1	PROC TTEST DATA=cash;
2	class School;
3	*may need to convert School to numeric;
4	var Money;
5	RUN;

3 Code Block

PROC IML Data

Explanation :
Using the IML matrix language to read data, generate 1000 random permutations of the 'Money' column (resampling without replacement) and create a large 'newds' table containing the original groups and the 1000 permuted vectors.

Copied!

1	ods OUTPUT off;
2	ods exclude all;
3
4	PROC IML ;
5	use cash;
6	read all var{School Money} into x;
7	p=t(ranperm(x[, 2], 1000));
8	paf=x[, 1]\|\|p;
9	create newds from paf;
10	append from paf;
11	QUIT;

4 Code Block

PROC TTEST Data

Explanation :
Massive execution of t-tests on the 1000 permuted columns (col2 to col1001) against the group (col1). Results (confidence limits/differences) are captured in the 'diff' table via ODS OUTPUT.

Copied!

1	ods OUTPUT conflimits=diff;
2
3	PROC TTEST DATA=newds plots=none;
4	class col1;
5	var col2 - col1001;
6	RUN;
7
8	ods OUTPUT on;
9	ods exclude none;

5 Code Block

PROC UNIVARIATE

Explanation :
Analysis of the distribution of mean differences (variable 'mean' in the output table 'diff') obtained by permutation, with histogram generation.

Copied!

1	PROC UNIVARIATE DATA=diff;
2	where method="Pooled";
3	var mean;
4	histogram mean;
5	RUN;

6 Code Block

DATA STEP Data

Explanation :
Filtering permutations whose mean difference is as extreme or more extreme than the observed difference (hardcoded here at 114.6). The resulting number of observations divided by 1000 gives the estimated p-value.

Copied!

1	DATA numdiffs;
2	SET diff;
3	where method="Pooled";
4
5	IF abs(mean) >=114.6;
6	RUN;
7
8	PROC PRINT DATA=numdiffs;
9	where method="Pooled";
10	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