Trailing-Edge
-
PDP-10 Archives
-
decuslib10-02
-
43,50145/qsf.doc
There are 2 other files named qsf.doc in the archive. Click here to see a list.
SUBROUTINE QSF
PURPOSE
TO COMPUTE THE VECTOR OF INTEGRAL VALUES FOR A GIVEN
EQUIDISTANT TABLE OF FUNCTION VALUES.
USAGE
CALL QSF (H,Y,Z,NDIM)
DESCRIPTION OF PARAMETERS
H - THE INCREMENT OF ARGUMENT VALUES.
Y - THE INPUT VECTOR OF FUNCTION VALUES.
Z - THE RESULTING VECTOR OF INTEGRAL VALUES. Z MAY BE
IDENTICAL WITH Y.
NDIM - THE DIMENSION OF VECTORS Y AND Z.
REMARKS
NO ACTION IN CASE NDIM LESS THAN 3.
SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
NONE
METHOD
BEGINNING WITH Z(1)=0, EVALUATION OF VECTOR Z IS DONE BY
MEANS OF SIMPSONS RULE TOGETHER WITH NEWTONS 3/8 RULE OR A
COMBINATION OF THESE TWO RULES. TRUNCATION ERROR IS OF
ORDER H**5 (I.E. FOURTH ORDER METHOD). ONLY IN CASE NDIM=3
TRUNCATION ERROR OF Z(2) IS OF ORDER H**4.
FOR REFERENCE, SEE
(1) F.B.HILDEBRAND, INTRODUCTION TO NUMERICAL ANALYSIS,
MCGRAW-HILL, NEW YORK/TORONTO/LONDON, 1956, PP.71-76.
(2) R.ZURMUEHL, PRAKTISCHE MATHEMATIK FUER INGENIEURE UND
PHYSIKER, SPRINGER, BERLIN/GOETTINGEN/HEIDELBERG, 1963,
PP.214-221.