Maximize a function given a table of values by spline interpolation.
Maximize a function given a table of values by spline interpolation.
maximizeInterpolant(x, y)
x |
numeric vector of the inputs of the function. |
y |
numeric matrix of function values at the values of |
Calculates the cubic spline interpolant for each row the method of Forsythe et al (1977) using the function fmm_spline from splines.c in the stats package).
Then calculates the derivatives of the spline segments adjacant to the input with the maximum function value.
This allows identification of the maximum of the interpolating spline.
numeric vector of input values at which the function maximums occur.
Aaron Lun, improving on earlier code by Gordon Smyth
Forsythe, G. E., Malcolm, M. A. and Moler, C. B. (1977). Computer Methods for Mathematical Computations, Prentice-Hall.
x <- seq(0,1,length=10) y <- rnorm(10,1,1) maximizeInterpolant(x,y)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.