uniform.hpp

Go to the documentation of this file.

#ifndef __STAN__PROB__DISTRIBUTIONS__UNIVARIATE__CONTINUOUS__UNIFORM_HPP__
#define __STAN__PROB__DISTRIBUTIONS__UNIVARIATE__CONTINUOUS__UNIFORM_HPP__

#include <boost/random/uniform_real_distribution.hpp>
#include <boost/random/variate_generator.hpp>

#include <stan/agrad.hpp>
#include <stan/math/error_handling.hpp>
#include <stan/math/functions/value_of.hpp>
#include <stan/meta/traits.hpp>
#include <stan/prob/constants.hpp>
#include <stan/prob/traits.hpp>

namespace stan {

  namespace prob {

    // CONTINUOUS, UNIVARIATE DENSITIES
    template <bool propto,
              typename T_y, typename T_low, typename T_high, 
              class Policy>
    typename return_type<T_y,T_low,T_high>::type
    uniform_log(const T_y& y, const T_low& alpha, const T_high& beta, 
                const Policy&) {
      static const char* function = "stan::prob::uniform_log(%1%)";
      
      using stan::math::check_not_nan;
      using stan::math::check_finite;
      using stan::math::check_greater;
      using stan::math::value_of;
      using stan::math::check_consistent_sizes;

      // check if any vectors are zero length
      if (!(stan::length(y) 
            && stan::length(alpha) 
            && stan::length(beta)))
        return 0.0;

      // set up return value accumulator
      double logp(0.0);
      if(!check_not_nan(function, y, "Random variable", &logp, Policy()))
        return logp;
      if (!check_finite(function, alpha, "Lower bound parameter", &logp, Policy()))
        return logp;
      if (!check_finite(function, beta, "Upper bound parameter", &logp, Policy()))
        return logp;
      if (!check_greater(function, beta, alpha, "Upper bound parameter",
                         &logp, Policy()))
        return logp;
      if (!(check_consistent_sizes(function,
                                   y,alpha,beta,
           "Random variable","Lower bound parameter","Upper bound parameter",
                                   &logp, Policy())))
        return logp;

      
      // check if no variables are involved and prop-to
      if (!include_summand<propto,T_y,T_low,T_high>::value)
  return 0.0;

      VectorView<const T_y> y_vec(y);
      VectorView<const T_low> alpha_vec(alpha);
      VectorView<const T_high> beta_vec(beta);
      size_t N = max_size(y, alpha, beta);

      for (size_t n = 0; n < N; n++) {
  const double y_dbl = value_of(y_vec[n]);
  if (y_dbl < value_of(alpha_vec[n]) || y_dbl > value_of(beta_vec[n]))
    return LOG_ZERO;
      }

      DoubleVectorView<include_summand<propto,T_low,T_high>::value,
  is_vector<T_low>::value | is_vector<T_high>::value> inv_beta_minus_alpha(max_size(alpha,beta));
      for (size_t i = 0; i < max_size(alpha,beta); i++) 
  if (include_summand<propto,T_low,T_high>::value)
    inv_beta_minus_alpha[i] = 1.0 / (value_of(beta_vec[i]) - value_of(alpha_vec[i]));
      DoubleVectorView<include_summand<propto,T_low,T_high>::value,
  is_vector<T_low>::value | is_vector<T_high>::value> log_beta_minus_alpha(max_size(alpha,beta));
      for (size_t i = 0; i < max_size(alpha,beta); i++)
  if (include_summand<propto,T_low,T_high>::value)
    log_beta_minus_alpha[i] = log(value_of(beta_vec[i]) - value_of(alpha_vec[i]));
      
      agrad::OperandsAndPartials<T_y,T_low,T_high> operands_and_partials(y,alpha,beta);
      for (size_t n = 0; n < N; n++) {
  if (include_summand<propto,T_low,T_high>::value)
    logp -= log_beta_minus_alpha[n];

  if (!is_constant_struct<T_low>::value)
    operands_and_partials.d_x2[n] += inv_beta_minus_alpha[n];
  if (!is_constant_struct<T_high>::value)
    operands_and_partials.d_x3[n] -= inv_beta_minus_alpha[n];
      }
      return operands_and_partials.to_var(logp);
    }


    template <bool propto,
              typename T_y, typename T_low, typename T_high>
    inline
    typename return_type<T_y,T_low,T_high>::type
    uniform_log(const T_y& y, const T_low& alpha, const T_high& beta) {
      return uniform_log<propto>(y,alpha,beta,stan::math::default_policy());
    }

    template <typename T_y, typename T_low, typename T_high, 
              class Policy>
    inline
    typename return_type<T_y,T_low,T_high>::type
    uniform_log(const T_y& y, const T_low& alpha, const T_high& beta, 
                const Policy&) {
      return uniform_log<false>(y,alpha,beta,Policy());
    }


    template <typename T_y, typename T_low, typename T_high>
    inline
    typename return_type<T_y,T_low,T_high>::type
    uniform_log(const T_y& y, const T_low& alpha, const T_high& beta) {
      return uniform_log<false>(y,alpha,beta,stan::math::default_policy());
    }

   
    template <class RNG>
    inline double
    uniform_rng(const double alpha,
                const double beta,
                RNG& rng) {
      using boost::variate_generator;
      using boost::random::uniform_real_distribution;
      variate_generator<RNG&, uniform_real_distribution<> >
        uniform_rng(rng, uniform_real_distribution<>(alpha, beta));
      return uniform_rng();
    }
  }
}
#endif