mcts::RunningStats Class Reference

Computes variance, mean and standart deviation. http://www.johndcook.com/standard_deviation.html http://www.johndcook.com/skewness_kurtosis.html. More...

#include <running_stats.hpp>

Public Member Functions
void	clear ()
void	push (const double &x)
	add a new value to the object More...
long long	num_data_values () const
	number of samples from which mean etc are calculated More...
double	mean () const
double	variance () const
double	standard_deviation () const
double	skewness () const
double	kurtosis () const
RunningStats &	operator+= (const RunningStats &rhs)

Friends
RunningStats	operator+ (const RunningStats a, const RunningStats b)

Detailed Description

Computes variance, mean and standart deviation. http://www.johndcook.com/standard_deviation.html http://www.johndcook.com/skewness_kurtosis.html.

Definition at line 12 of file running_stats.hpp.

Member Function Documentation

num_data_values()

long long mcts::RunningStats::num_data_values

(

)

const

number of samples from which mean etc are calculated

Returns

number of samples

Definition at line 30 of file running_stats.cpp.

{ return n; }

Referenced by mcts::AvgBackpropagationStrategy::nb_samples().

push()

void mcts::RunningStats::push

(

const double &

x

)

add a new value to the object

Parameters

x	numeric input value

Definition at line 14 of file running_stats.cpp.

                                       {
  double delta, delta_n, delta_n2, term1;

  long long n1 = n;
  n++;
  delta = x - M1;
  delta_n = delta / n;
  delta_n2 = delta_n * delta_n;
  term1 = delta * delta_n * n1;
  M1 += delta_n;
  M4 += term1 * delta_n2 * (n * n - 3 * n + 3) + 6 * delta_n2 * M2 -
        4 * delta_n * M3;
  M3 += term1 * delta_n * (n - 2) - 3 * delta_n * M2;
  M2 += term1;
}

Referenced by mcts::AvgBackpropagationStrategy::on_backpropagate().

---

The documentation for this class was generated from the following files:

• include/mcts/running_stats.hpp
• src/running_stats.cpp