scicpp::stats::var, nanvar, tvar¶

Defined in header <scicpp/core.hpp>

Compute the variance. ddof is the delta degrees of freedom

template<int ddof = 0, class Array, class Predicate>
constexpr auto var(const Array &f, Predicate filter)¶

Compute the variance of an array using only the values satisfying the filter predicate.

template<int ddof = 0, class Array>
constexpr auto var(const Array &f)¶

Compute the variance of an array.

template<int ddof = 0, class Array>
auto nanvar(const Array &f)¶

Compute the variance of an array, ignoring NaNs.

template<int ddof = 1, class Array>
constexpr auto tvar(const Array &f, const std::array<T, 2> &limits, const std::array<bool, 2> &inclusive)¶

Compute the variance of an array, ignoring values outside the given limits. Use the optional parameter inclusive to specify whether the limits are included (by default they are included). Note that by default here ddof = 1.

#include <cstdio>
#include <scicpp/core.hpp>

int main()
{
    const std::array a{-1., 1., 2., 3.};

    // Compute the variance of the array
    const auto v = scicpp::stats::var(a);

    // Compute the variance of the positive values of the array
    const auto v_pos = scicpp::stats::var(a, [](auto x) { return x >= 0; });

    printf("v = %f, v_pos = %f\n", v, v_pos);
}