source: MML/trunk/machine/SOLEIL/common/toolbox/PolyfitnTools/PolyfitnTools/polyvaln.m @ 4

function ypred = polyvaln(polymodel,indepvar)
% polyvaln: evaluates a polynomial model as a function of its variables
% usage: ypred = polyvaln(polymodel,indepvar)
%
% arguments: (input)
%  indepvar - (n x p) array of independent variables as columns
%        n is the number of data points to evaluate
%        p is the dimension of the independent variable space
%
%        IF n == 1, then I will assume there is only a
%        single independent variable.
%
%  polymodel - A structure containing a regression model from polyfitn
%        polymodel.ModelTerms = list of terms in the model
%        polymodel.Coefficients = regression coefficients
%  
%        Note: A polymodel can be evaluated for any set of
%        values with the function polyvaln. However, if you
%        wish to manipulate the result symbolically using my
%        own sympoly tools, this structure should be converted
%        to a sympoly using the function polyn2sympoly.
%
% Arguments: (output)
%  ypred - nx1 vector of predictions through the model.
%
%
% See also: polyfitn, polyfit, polyval, polyn2sympoly, sympoly
%
% Author: John D'Errico
% Release: 1.0
% Release date: 2/19/06

% get the size of indepvar
[n,p] = size(indepvar);
if (n == 1) && (size(polymodel.ModelTerms,2)==1)
  indepvar = indepvar';
  [n,p] = size(indepvar);
elseif (size(polymodel.ModelTerms,2)~=p)
  error 'Size of indepvar array and this model are inconsistent.'
end

% Evaluate the model
nt = size(polymodel.ModelTerms,1);
ypred = zeros(n,1);
for i = 1:nt
  t = ones(n,1);
  for j = 1:p
    t = t.*indepvar(:,j).^polymodel.ModelTerms(i,j);
  end
  ypred = ypred + t*polymodel.Coefficients(i);
end