SUBROUTINE EXPRAN(N,ISEED,X)
C
C     PURPOSE--THIS SUBROUTINE GENERATES A RANDOM SAMPLE OF SIZE N
C              FROM THE EXPONENTIAL DISTRIBUTION
C              WITH MEAN = 1 AND STANDARD DEVIATION = 1.
C              THIS DISTRIBUTION IS DEFINED FOR ALL NON-NEGATIVE X,
C              AND HAS THE PROBABILITY DENSITY FUNCTION
C              F(X) = EXP(-X).
C     INPUT  ARGUMENTS--N      = THE DESIRED INTEGER NUMBER
C                                OF RANDOM NUMBERS TO BE
C                                GENERATED.
C     OUTPUT ARGUMENTS--X      = A SINGLE PRECISION VECTOR
C                                (OF DIMENSION AT LEAST N)
C                                INTO WHICH THE GENERATED
C                                RANDOM SAMPLE WILL BE PLACED.
C     OUTPUT--A RANDOM SAMPLE OF SIZE N
C             FROM THE EXPONENTIAL DISTRIBUTION
C             WITH MEAN = 1 AND STANDARD DEVIATION = 1.
C     PRINTING--NONE UNLESS AN INPUT ARGUMENT ERROR CONDITION EXISTS.
C     RESTRICTIONS--THERE IS NO RESTRICTION ON THE MAXIMUM VALUE
C                   OF N FOR THIS SUBROUTINE.
C     OTHER DATAPAC   SUBROUTINES NEEDED--UNIRAN.
C     FORTRAN LIBRARY SUBROUTINES NEEDED--ALOG.
C     MODE OF INTERNAL OPERATIONS--SINGLE PRECISION.
C     LANGUAGE--ANSI FORTRAN (1977)
C     REFERENCES--TOCHER, THE ART OF SIMULATION,
C                 1963, PAGES 14, 35-36.
C               --HAMMERSLEY AND HANDSCOMB, MONTE CARLO METHODS,
C                 1964, PAGE 36.
C               --FILLIBEN, 'THE PERCENT POINT FUNCTION',
C                 (UNPUBLISHED MANUSCRIPT), 1970, PAGES 28-31.
C               --JOHNSON AND KOTZ, CONTINUOUS UNIVARIATE
C                 DISTRIBUTIONS--1, 1970, PAGES 207-232.
C               --HASTINGS AND PEACOCK, STATISTICAL
C                 DISTRIBUTIONS--A HANDBOOK FOR
C                 STUDENTS AND PRACTITIONERS, 1975,
C                 PAGE 58.
C     WRITTEN BY--JAMES J. FILLIBEN
C                 STATISTICAL ENGINEERING DIVISION
C                 CENTER FOR APPLIED MATHEMATICS
C                 NATIONAL BUREAU OF STANDARDS
C                 WASHINGTON, D. C. 20234
C                 PHONE--301-921-3651
C     NOTE--DATAPLOT IS A REGISTERED TRADEMARK
C           OF THE NATIONAL BUREAU OF STANDARDS.
C           THIS SUBROUTINE MAY NOT BE COPIED, EXTRACTED,
C           MODIFIED, OR OTHERWISE USED IN A CONTEXT
C           OUTSIDE OF THE DATAPLOT LANGUAGE/SYSTEM.
C     LANGUAGE--ANSI FORTRAN (1966)
C               EXCEPTION--HOLLERITH STRINGS IN FORMAT STATEMENTS
C                          DENOTED BY QUOTES RATHER THAN NH.
C     VERSION NUMBER--82/7
C     ORIGINAL VERSION--JUNE      1972.
C     UPDATED         --SEPTEMBER 1975.
C     UPDATED         --NOVEMBER  1975.
C     UPDATED         --JULY      1976.
C     UPDATED         --DECEMBER  1981.
C     UPDATED         --MAY       1982.
C
C-----CHARACTER STATEMENTS FOR NON-COMMON VARIABLES-------------------
C
C---------------------------------------------------------------------
C
      DIMENSION X(*)
C
C---------------------------------------------------------------------
C
CCCCC CHARACTER*4 IFEEDB
CCCCC CHARACTER*4 IPRINT
C
CCCCC COMMON /MACH/IRD,IPR,CPUMIN,CPUMAX,NUMBPC,NUMCPW,NUMBPW
CCCCC COMMON /PRINT/IFEEDB,IPRINT
C
      IPR=6
C
C-----START POINT-----------------------------------------------------
C
C     CHECK THE INPUT ARGUMENTS FOR ERRORS
C
      IF(N.LT.1)GOTO50
      GOTO90
   50 WRITE(IPR, 5)
      WRITE(IPR,47)N
      RETURN
   90 CONTINUE
    5 FORMAT(1H , 91H***** FATAL ERROR--THE FIRST  INPUT ARGUMENT TO THE
     1 EXPRAN SUBROUTINE IS NON-POSITIVE *****)
   47 FORMAT(1H , 35H***** THE VALUE OF THE ARGUMENT IS ,I8   ,6H *****)
C
C     GENERATE N UNIFORM (0,1) RANDOM NUMBERS;
C
      CALL UNIRAN(N,ISEED,X)
C
C     GENERATE N EXPONENTIAL RANDOM NUMBERS
C     USING THE PERCENT POINT FUNCTION TRANSFORMATION METHOD.
C
      DO100I=1,N
      X(I)=-ALOG(X(I))
  100 CONTINUE
C
      RETURN
      END