4

I am looking for a way to compute the linear interpolation of a multivariate function (of 5 variables) using R. The package akima offer a way to compute the linear interpolation in the bivariate case. Is possible to perform the multivariate linear interpolation in matlab using the function interp, but I was wondering if there is a way to do the same in R.

Hope someone can help me! Thanks!

Sabrina Vettori
  • 41
  • 2
  • You may have to do bivariate interp, then apply the results to the next pair of variables, and so on. – Carl Witthoft Mar 12 '15 at 12:41
  • Thanks but I am not sure I am understanding how to do this. Could be more clear please? I have the values of my function "A" (vector) evaluated at the corresponding points "xyzgh" (matrix of 5 columns). Shall I apply A1 <- interp1(x, y, A, method = "linear") and then? – Sabrina Vettori Mar 12 '15 at 12:50
  • possible duplicate of http://stackoverflow.com/questions/7142180/is-there-an-r-library-that-estimates-a-multivariate-natural-cubic-spline-or-sim?rq=1 – Carl Witthoft Mar 12 '15 at 19:06

1 Answers1

2

interp now provides this within r: https://rdrr.io/cran/npsp/man/interp.html & https://cran.r-project.org/web/packages/interp/interp.pdf

See also chebpol https://cran.r-project.org/web/packages/chebpol/chebpol.pdf

dez93_2000
  • 1,730
  • 2
  • 23
  • 34