Example 2.3 Analysis Using Fisher’s z Transformation
The following statements request Pearson correlation statistics by using Fisher’s
transformation for the data set Fitness:
proc corr data=Fitness nosimple fisher;
var weight oxygen runtime;
run;
The NOSIMPLE option suppresses the table of univariate descriptive statistics. By default, PROC CORR displays the "Pearson Correlation Coefficients" table in Output 2.3.1.
Output 2.3.1
Pearson Correlations
Using the FISHER option, the CORR procedure displays correlation statistics by using Fisher’s
transformation in Output 2.3.2.
Output 2.3.2
Correlation Statistics Using Fisher’s
Transformation
29 |
-0.15358 |
-0.15480 |
-0.00274 |
-0.15090 |
-0.490289 |
0.228229 |
0.4299 |
29 |
0.20072 |
0.20348 |
0.00358 |
0.19727 |
-0.182422 |
0.525765 |
0.2995 |
28 |
-0.86843 |
-1.32665 |
-0.01608 |
-0.86442 |
-0.935728 |
-0.725221 |
<.0001 |
The table also displays confidence limits and a
-value for the default null hypothesis
. See the section Fisher’s z Transformation for details on Fisher’s
transformation.
The following statements request one-sided hypothesis tests and confidence limits for the correlations using Fisher’s
transformation:
proc corr data=Fitness nosimple nocorr fisher (type=lower);
var weight oxygen runtime;
run;
The NOSIMPLE option suppresses the "Simple Statistics" table, and the NOCORR option suppresses the "Pearson Correlation Coefficients" table.
Output 2.3.3 displays correlation statistics by using Fisher’s
transformation.
Output 2.3.3
One-Sided Correlation Analysis Using Fisher’s
Transformation
29 |
-0.15358 |
-0.15480 |
-0.00274 |
-0.15090 |
-0.441943 |
0.7850 |
29 |
0.20072 |
0.20348 |
0.00358 |
0.19727 |
-0.122077 |
0.1497 |
28 |
-0.86843 |
-1.32665 |
-0.01608 |
-0.86442 |
-0.927408 |
1.0000 |
The FISHER(TYPE=LOWER) option requests a lower confidence limit and a
-value for the test of the one-sided hypothesis
against the alternative hypothesis
. Here Fisher’s
, the bias adjustment, and the estimate of the correlation are the same as for the two-sided alternative. However, because TYPE=LOWER is specified, only a lower confidence limit is computed for each correlation, and one-sided
-values are computed.
Copyright © SAS Institute, Inc. All Rights Reserved.