}

  static class MeanAndVariance {
    private final double mean;
    private final double variance;

    MeanAndVariance(double mean, double variance) {
      this.mean = mean;
      this.variance = variance;
    }

    @Override
    public int hashCode() {
      return Doubles.hashCode(mean) * 31 + Doubles.hashCode(variance);
    }
  }

  enum VarianceAlgorithm {
    DO_NOT_COMPUTE {
      @Override

   * either the {@code x} or {@code y} data must have a non-zero population variance (i.e. {@code
   * xStats().populationVariance() > 0.0 || yStats().populationVariance() > 0.0}). The result is
   * guaranteed to be horizontal if there is variance in the {@code x} data but not the {@code y}
   * data, and vertical if there is variance in the {@code y} data but not the {@code x} data.
   *

   * either the {@code x} or {@code y} data must have a non-zero population variance (i.e. {@code
   * xStats().populationVariance() > 0.0 || yStats().populationVariance() > 0.0}). The result is
   * guaranteed to be horizontal if there is variance in the {@code x} data but not the {@code y}
   * data, and vertical if there is variance in the {@code y} data but not the {@code x} data.
   *

   * either the {@code x} or {@code y} data must have a non-zero population variance (i.e. {@code
   * xStats().populationVariance() > 0.0 || yStats().populationVariance() > 0.0}). The result is
   * guaranteed to be horizontal if there is variance in the {@code x} data but not the {@code y}
   * data, and vertical if there is variance in the {@code y} data but not the {@code x} data.
   *

   * either the {@code x} or {@code y} data must have a non-zero population variance (i.e. {@code
   * xStats().populationVariance() > 0.0 || yStats().populationVariance() > 0.0}). The result is
   * guaranteed to be horizontal if there is variance in the {@code x} data but not the {@code y}
   * data, and vertical if there is variance in the {@code y} data but not the {@code x} data.
   *

    return Math.sqrt(populationVariance());
  }

  /**
   * Returns the <a href="http://en.wikipedia.org/wiki/Variance#Sample_variance">unbiased sample
   * variance</a> of the values. If this dataset is a sample drawn from a population, this is an
   * unbiased estimator of the population variance of the population. The count must be greater than
   * one.
   *

    return Math.sqrt(populationVariance());
  }

  /**
   * Returns the <a href="http://en.wikipedia.org/wiki/Variance#Sample_variance">unbiased sample
   * variance</a> of the values. If this dataset is a sample drawn from a population, this is an
   * unbiased estimator of the population variance of the population. The count must be greater than
   * one.
   *

      if (values.hasAnyNonFinite()) {
        assertWithMessage("population variance of " + values).that(populationVariance).isNaN();
        assertWithMessage("population variance by addAll(Stats) of " + values)
            .that(populationVarianceByAddAllStats)
            .isNaN();
      } else {
        assertWithMessage("population variance of " + values)
            .that(populationVariance)
            .isWithin(ALLOWED_ERROR)

      if (values.hasAnyNonFinite()) {
        assertWithMessage("population variance of " + values).that(populationVariance).isNaN();
        assertWithMessage("population variance by addAll(Stats) of " + values)
            .that(populationVarianceByAddAllStats)
            .isNaN();
      } else {
        assertWithMessage("population variance of " + values)
            .that(populationVariance)
            .isWithin(ALLOWED_ERROR)

      double populationVariance = Stats.of(values.asIterable()).populationVariance();
      if (values.hasAnyNonFinite()) {
        assertWithMessage("population variance of " + values).that(populationVariance).isNaN();
      } else {
        assertWithMessage("population variance of " + values)
            .that(populationVariance)
            .isWithin(ALLOWED_ERROR)
            .of(MANY_VALUES_SUM_OF_SQUARES_OF_DELTAS / MANY_VALUES_COUNT);