 
      SUBROUTINE QUICK(MM, M, N, A, CLAB, RLAB, TITLE, THRESH, XMISS,
     *                 NC, IWORK, OUNIT)
C
C<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
C
C   PURPOSE
C   -------
C
C      FINDS A QUICK PARTITION OF THE CASES BY COMPARING, TO A USER-
C      DEFINED THRESHOLD, THE EUCLIDEAN DISTANCES TO THE EXISTING
C      CLUSTER LEADERS
C
C   DESCRIPTION
C   -----------
C
C   1.  INITIALLY, THE FIRST CASE WILL BE ASSIGNED TO THE FIRST CLUSTER
C       AND BECOMES THE LEADER OF THE FIRST CLUSTER.  THEN, GIVEN A NEW
C       CASE, CYCLE THROUGH THE EXISTING CLUSTERS IN ORDER.  PLACE THE
C       CASE IN THE FIRST CLUSTER WHERE THE DISTANCE BETWEEN THE CASE
C       AND THE CLUSTER LEADER IS LESS THAN THE THRESHOLD.  IF NO
C       CLUSTER EXISTS, PLACE THE CASE IN A NEW CLUSTER MAKING IT THE
C       CLUSTER LEADER.  ONCE THE MAXIMUM NUMBER OF DESIRED CLUSTERS
C       HAS BEEN REACHED, NO NEW CLUSTERS WILL BE FORMED AND CASES NOT
C       BELONGING TO AN EXISTING CLUSTER WILL BE IGNORED.
C
C   2.  THE DISTANCE FUNCTION USED IS THE EUCLIDEAN DISTANCE.  THE
C       VARIABLES SHOULD BE SCALED SIMILARLY (CLUSTER SUBROUTINE STAND
C       CAN BE USED TO STANDARDIZE THE VARIABLES).  ANY MISSING VALUES
C       WILL BE IGNORED IN THE DISTANCE CALCULATION.
C
C   3.  THE OUTPUT IS ON FORTRAN UNIT OUNIT, WHICH FOR EACH CLUSTER IS
C       THE CLUSTER LEADER AND ITS VALUES FOLLOWED BY THE OTHER
C       MEMBERS.
C
C   INPUT PARAMETERS
C   ----------------
C
C   MM    INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE FIRST DIMENSION OF THE MATRIX A.  MUST BE AT LEAST M.
C
C   M     INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE NUMBER OF CASES.
C
C   N     INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE NUMBER OF VARIABLES.
C
C   A     REAL MATRIX WHOSE FIRST DIMENSION MUST BE MM AND WHOSE SECOND
C            DIMENSION MUST BE AT LEAST N (UNCHANGED ON OUTPUT).
C         THE MATRIX OF DATA VALUES.
C
C         A(I,J) IS THE VALUE FOR THE J-TH VARIABLE FOR THE I-TH CASE.
C
C   CLAB  VECTOR OF 4-CHARACTER VARIABLES DIMENSIONED AT LEAST N.
C            (UNCHANGED ON OUTPUT).
C         THE LABELS OF THE VARIABLES.
C
C   RLAB  VECTOR OF 4-CHARACTER VARIABLES DIMENSIONED AT LEAST M.
C            (UNCHANGED ON OUTPUT).
C         THE LABELS OF THE CASES.
C
C   TITLE 10-CHARACTER VARIABLE (UNCHANGED ON OUTPUT).
C         TITLE OF THE DATA SET.
C
C   THRESH REAL SCALAR (UNCHANGED ON OUTPUT).
C         THRESHOLD SUCH THAT ANY TWO CASES WHOSE DISTANCE IS LESS
C         THAN THRESH WILL BE ASSIGNED TO THE SAME CLUSTER.
C
C   XMISS REAL SCALAR (UNCHANGED ON OUTPUT).
C         MISSING VALUE CODE.  IF A(I,J) = XMISS, THEN THE VALUE FOR THE
C         J-TH VARIABLE FOR THE I-TH CASE IS ASSUMED TO BE MISSING.
C
C   NC    INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         MAXIMUM NUMBER OF CLUSTERS DESIRED.
C
C   IWORK INTEGER VECTOR DIMENSIONED AT LEAST M+NC.
C         WORK VECTOR.
C
C   OUNIT INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         UNIT NUMBER FOR OUTPUT.
C
C   REFERENCE
C   ---------
C
C     HARTIGAN, J. A. (1975).  CLUSTERING ALGORITHMS, JOHN WILEY &
C        SONS, INC., NEW YORK.  PAGES 74-83.
C
C<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
C
 
 
