dummycode.m

The BNL toolbox is a set of Matlab functions for defining and estimating the parameters of a Bayesi

function y=dummycode(x,baselinecategory)
%y=dummycode(x,baselinecategory) recodes categorical variable (0,1,...k) into k-1 dummies
% x is a vector, y is a lenght(x)*(k-1) matrix
if nargin==1 baselinecategory=0;,end;

if ~isvector(x)
     error('dummycode takes only vectors as inputs'),return, end;
m=min(x);
r=range(x);
a= find(m==0);
if isempty(a) error('no observations in category 0'),return,end;
y=zeros(length(x), r-1);
for i=1:r+1
    a=find(x==(i-1));
    if isempty(a) error('at least one empty category'),return,end;
    y(a,i)=1;
end
y(:,baselinecategory+1)=[];