Google
 

Trailing-Edge - PDP-10 Archives - decuslib20-02 - decus/20-0026/stout.doc
There are 2 other files named stout.doc in the archive. Click here to see a list.
SAMPLE OUTPUT SUBROUTINE STOUT

PURPOSE
   PRINT THE RESULT OF A STEP-WISE MULTIPLE REGRESSION.  THIS
   SUBROUTINE IS CALLED BY THE SUBROUTINE STPRG.

USAGE
   CALL STOUT (NSTEP,ANS,L,B,S,T,NSTOP)

DESCRIPTION OF PARAMETERS
   NSTEP - INPUT VECTOR OF LENGTH 5 CONTAINING THE FOLLOWING
	   INFORMATION..
	     NSTEP(1)  DEPENDENT VARIABLE
	     NSTEP(2)  NUMBER OF VARIABLES FORCED TO ENTER
		       IN THE REGRESSION
	     NSTEP(3)  NUMBER OF VARIABLES DELETED
	     NSTEP(4)  THE LAST STEP NUMBER
	     NSTEP(5)  THE LAST VARIABLE ENTERED
   ANS	 - INPUT VECTOR OF LENGTH 11 CONTAINING THE FOLLOWING
	   INFORMATION FOR THE LAST STEP..
	     ANS(1)  SUM OF SQUARES REDUCED
	     ANS(2)  PROPORTION REDUCED
	     ANS(3)  CUMULATIVE SUM OF SQUARES REDUCED
	     ANS(4)  CUMULATIVE PROPORTION REDUCED
	     ANS(5)  SUM OF SQUARES OF THE DEPENDENT VARIABLE
	     ANS(6)  MULTIPLE CORRELATION COEFFICIENT
	     ANS(7)  F-VALUE FOR ANALYSIS VARIANCE (FOR THE
		     REGRESSION)
	     ANS(8)  STANDARD ERROR OF ESTIMATE
	     ANS(9)  INTERCEPT
	     ANS(10) ADJUSTED MULTIPLE R
	     ANS(11) ADJUSTED STANDARD ERROR OF ESTIMATE
   L	 - INPUT VECTOR OF LENGTH K (K=M-NSTEP(3)-1) CONTAIN-
	   ING VARIABLES ENTERED IN THE REGRESSION.  L(1)=FIRST
	   VARIABLE ENTERED, L(2)=SECOND VARIABLE ENTERED, ETC.
   B	 - INPUT VECTOR OF LENGTH K (K=M-NSTEP(3)-1) CONTAIN-
	   ING REGRESSION COEFFICIENTS CORRESPONDING TO THE
	   VARIABLES IN VECTOR L
   S	 - INPUT VECTOR OF LENGTH K (K=M-NSTEP(3)-1) CONTAIN-
	   ING STANDARD ERRORS OF REGRESSION COEFFICIENTS
	   CORRESPONDING TO THE VARIABLES IN VECTOR L
   T	 - INPUT VECTOR OF LENGTH K (K=M-NSTEP(3)-1) CONTAIN-
	   ING COMPUTED T-VALUES CORRESPONDING TO THE VARIABLES
	   IN VECTOR L
   NSTOP - OUTPUT OPTION CODE TO STOP THE STEP-WISE REGRESSION
	     1 - IF THE STEP-WISE REGRESSION IS TO BE TERMI-
		 NATED BY SOME CRITERIA OTHER THAN PROPORTION
		 OF SUM OF SQUARES, SUCH AS F-TEST AND SO ON,
		 THIS SUBROUTINE MAY BE MODIFIED TO PERFORM
		 DESIRED TESTS.  WHEN IT BECOMES NO LONGER
		 NECESSARY TO CONTINUE THE STEP-WISE REGRES-
		 SION, SET NSTOP EQUAL TO 1.
	     0 - IF THE STEP-WISE REGRESSION IS TO BE CONTINUED

REMARKS
   THE CONTENTS OF THE VECTORS NSTEP, ANS, L ARE REQUIRED IN
   SUBSEQUENT STEPS AND MUST NOT BE DESTROYED.

SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
   NONE