Previous: fitdsm Up: ../plot79_f.html Next: fitk1
SUBROUTINE FITIN (X, Y, L, U, V, N)
C$ (Interpolation of a Single Function)
C$ This subroutine interpolates, from values of the function
C$ given as ordinates of input data points in an X-Y plane and
C$ for a given set of X values (abscissas), the values of a
C$ single-valued function Y = Y(X).
C$
C$ The input arguments are:
C$
C$ X = Array of dimension L storing the X values (abscissas)
C$ of input data points (in ascending order)
C$ Y = Array of dimension L storing the Y values (ordinates)
C$ of input data points
C$ L = Number of input data points (must be 2 or greater)
C$ U = Array of dimension N storing the X values (abscissas)
C$ of desired points
C$ N = Number of points at which interpolation of the Y value
C$ (ordinate) is desired (must be 1 or greater)
C$
C$ The output argument is:
C$
C$ V = Array of dimension N where the interpolated Y
C$ values (ordinates) are to be displayed
C$
C$ Author: Hiroshi Akima, "Interpolation and Smooth Curve
C$ Fitting Based on Local Procedures", Comm. ACM 15,
C$ 914-918 (1972), and "A New Method of Interpolation
C$ and Smooth Curve Fitting Based on Local
C$ Procedures", J. ACM 17, 589-602 (1970).
C$
C$ Corrections: M.R. Andersen, "Remark on Algorithm 433",
C$ ACM Trans. on Math. Software, 2, 208 (1976).
C$
C$ Additional corrections: N.H.F. Beebe, "Remark on Algorithm
C$ 433", ACM Trans. on Math. Software,
C$ to be published (198x).
C$
C$ This routine has been subjected to the University of
C$ Colorado DAVE Global Data Flow Analysis System, and all
C$ data paths leading to possible use of uninitialized
C$ variables have been shown to be impossible paths. (N.H.F.
C$ Beebe, University of Utah, 22-NOV-80)
C$ (25-SEP-84)