29.12.2013 Views

Base SAS 9.1.3 Procedures Guide - Acsu Buffalo

Base SAS 9.1.3 Procedures Guide - Acsu Buffalo

Base SAS 9.1.3 Procedures Guide - Acsu Buffalo

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

PROC CORR Statement<br />

<br />

9<br />

CSSCP<br />

displays a table of the corrected sums of squares and crossproducts. When you specify<br />

the CSSCP option, the Pearson correlations will also be displayed. If you specify<br />

the OUTP= option, the output data set also contains a CSSCP matrix with the corresponding<br />

–TYPE– variable value ‘CSSCP.’ If you use a PARTIAL statement, PROC<br />

CORR prints both an unpartial and a partial CSSCP matrix, and the output data set<br />

contains a partial CSSCP matrix.<br />

DATA=<strong>SAS</strong>-data-set<br />

names the <strong>SAS</strong> data set to be analyzed by PROC CORR. By default, the procedure<br />

uses the most recently created <strong>SAS</strong> data set.<br />

EXCLNPWGT<br />

excludes observations with nonpositive weight values from the analysis. By default,<br />

PROC CORR treats observations with negative weights like those with zero weights<br />

and counts them in the total number of observations.<br />

FISHER < ( fisher-options ) ><br />

requests confidence limits and p-values under a specified null hypothesis, H 0 : ρ =<br />

ρ 0 , for correlation coefficients using Fisher’s z transformation. These correlations<br />

include the Pearson correlations and Spearman correlations.<br />

The following fisher-options are available:<br />

ALPHA= α<br />

specifies the level of the confidence limits for the correlation, 100(1 − α)%.<br />

The value of the ALPHA= option must be between 0 and 1, and the default is<br />

ALPHA=0.05.<br />

BIASADJ= YES | NO<br />

specifies whether or not the bias adjustment is used in constructing confidence<br />

limits. The BIASADJ=YES option also produces a new correlation estimate<br />

using the bias adjustment. By default, BIASADJ=YES.<br />

RHO0= ρ 0<br />

specifies the value ρ 0 in the null hypothesis H 0 : ρ = ρ 0 , where −1 < ρ 0 < 1.<br />

By default, RHO0=0.<br />

TYPE= LOWER | UPPER | TWOSIDED<br />

specifies the type of confidence limits. The TYPE=LOWER option requests<br />

a lower confidence limit from the lower alternative H 1 : ρ < ρ 0 , the<br />

TYPE=UPPER option requests an upper confidence limit from the upper alternative<br />

H 1 : ρ > ρ 0 , and the default TYPE=TWOSIDED option requests twosided<br />

confidence limits from the two-sided alternative H 1 : ρ ≠ ρ 0 .<br />

HOEFFDING<br />

requests a table of Hoeffding’s D statistics. This D statistic is 30 times larger than the<br />

usual definition and scales the range between −0.5 and 1 so that large positive values<br />

indicate dependence. The HOEFFDING option is invalid if a WEIGHT or PARTIAL<br />

statement is used.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!