function bounds = prior_bounds(bayestopt)

% function bounds = prior_bounds(bayestopt)
% bounds of the prior density
%
% INPUTS
%    bayestopt:        structure characterizing priors (shape, mean, p1..p4)
%    
% OUTPUTS
%    bounds:           matrix specifying bounds (raw= parameter, column=upper&lower bound)
%    
% SPECIAL REQUIREMENTS
%    none
%  
% part of DYNARE, copyright Dynare Team (2003-2008)
% Gnu Public License.


global options_

pshape = bayestopt.pshape;
pmean = bayestopt.pmean;
p1 = bayestopt.p1;
p2 = bayestopt.p2;
p3 = bayestopt.p3;
p4 = bayestopt.p4;

n = length(pmean);
bounds = zeros(n,2);

for i=1:n
  switch pshape(i)
    case 1
      mu = (pmean(i)-p3(i))/(p4(i)-p3(i));
      stdd = p2(i)/(p4(i)-p3(i));
      A = (1-mu)*mu^2/stdd^2 - mu;
      B = A*(1/mu - 1);
      bounds(i,1) = qbeta(options_.prior_trunc,A,B)*(p4(i)-p3(i))+p3(i);
      bounds(i,2) = qbeta(1-options_.prior_trunc,A,B)*(p4(i)-p3(i))+p3(i);
    case 2
      b = p2(i)^2/(pmean(i)-p3(i));
      a = (pmean(i)-p3(i))/b;
      bounds(i,1) = mj_qgamma(options_.prior_trunc,a)*b+p3(i);
      bounds(i,2) = mj_qgamma(1-options_.prior_trunc,a)*b+p3(i);
    case 3
      bounds(i,1) = qnorm(options_.prior_trunc,pmean(i),p2(i));
      bounds(i,2) = qnorm(1-options_.prior_trunc,pmean(i),p2(i));
    case 4
      nu = p2(i);
      mu = pmean(i);
      beta = ( gamma( (nu-1)/2 ) / mu / gamma( nu/2 ) )^2;
      a=2/beta;
      bounds(i,1) = 1/sqrt(mj_qgamma(1-options_.prior_trunc,p2(i)/2)*beta);
      bounds(i,2) = 1/sqrt(mj_qgamma(options_.prior_trunc,p2(i)/2)*beta);
    case 5
      bounds(i,1) = p1(i);
      bounds(i,2) = p2(i);
    otherwise
      bounds(i,1) = -Inf;
      bounds(i,2) = Inf;
  end
end