stat_utils_test.cc

Go to the documentation of this file.

 
/*
                          Aleph_w
 
  Data structures & Algorithms
  version 2.0.0b
  https://github.com/lrleon/Aleph-w
 
  This file is part of Aleph-w library
 
  Copyright (c) 2002-2026 Leandro Rabindranath Leon
 
  Permission is hereby granted, free of charge, to any person obtaining a copy
  of this software and associated documentation files (the "Software"), to deal
  in the Software without restriction, including without limitation the rights
  to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  copies of the Software, and to permit persons to whom the Software is
  furnished to do so, subject to the following conditions:
 
  The above copyright notice and this permission notice shall be included in all
  copies or substantial portions of the Software.
 
  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  SOFTWARE.
*/
 
 
#include <gtest/gtest.h>
#include <vector>
#include <array>
#include <list>
#include <cmath>
#include <stat_utils.H>
 
using namespace std;
using namespace Aleph;
 
// =============================================================================
// Sum Tests
// =============================================================================
 
TEST(SumTest, EmptyVector)
{
  vector<double> v;
  EXPECT_DOUBLE_EQ(sum(v), 0.0);
}
 
TEST(SumTest, SingleElement)
{
  vector<double> v = {5.0};
  EXPECT_DOUBLE_EQ(sum(v), 5.0);
}
 
TEST(SumTest, MultipleElements)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(sum(v), 15.0);
}
 
TEST(SumTest, NegativeNumbers)
{
  vector<double> v = {-1.0, -2.0, 3.0};
  EXPECT_DOUBLE_EQ(sum(v), 0.0);
}
 
TEST(SumTest, IntegerVector)
{
  vector<int> v = {1, 2, 3, 4, 5};
  EXPECT_EQ(sum(v), 15);
}
 
// =============================================================================
// Mean Tests
// =============================================================================
 
TEST(MeanTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(mean(v), invalid_argument);
}
 
TEST(MeanTest, SingleElement)
{
  vector<double> v = {5.0};
  EXPECT_DOUBLE_EQ(mean(v), 5.0);
}
 
TEST(MeanTest, MultipleElements)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(mean(v), 3.0);
}
 
TEST(MeanTest, NegativeNumbers)
{
  vector<double> v = {-10.0, 10.0};
  EXPECT_DOUBLE_EQ(mean(v), 0.0);
}
 
TEST(MeanTest, WithList)
{
  list<double> l = {1.0, 2.0, 3.0};
  EXPECT_DOUBLE_EQ(mean(l), 2.0);
}
 
// =============================================================================
// Variance Tests
// =============================================================================
 
TEST(VarianceTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(variance(v), invalid_argument);
}
 
TEST(VarianceTest, SingleElementThrows)
{
  vector<double> v = {5.0};
  EXPECT_THROW(variance(v, false), invalid_argument);  // Sample variance
}
 
TEST(VarianceTest, SingleElementPopulation)
{
  vector<double> v = {5.0};
  EXPECT_DOUBLE_EQ(variance(v, true), 0.0);  // Population variance
}
 
TEST(VarianceTest, TwoElements)
{
  vector<double> v = {0.0, 2.0};
  // Sample variance: ((0-1)^2 + (2-1)^2) / 1 = 2
  EXPECT_DOUBLE_EQ(variance(v, false), 2.0);
}
 
TEST(VarianceTest, SampleVsPopulation)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  double sample_var = variance(v, false);
  double pop_var = variance(v, true);
  // Sample variance should be larger (n-1 divisor vs n)
  EXPECT_GT(sample_var, pop_var);
}
 
TEST(VarianceTest, ConstantValues)
{
  vector<double> v = {5.0, 5.0, 5.0, 5.0};
  EXPECT_DOUBLE_EQ(variance(v), 0.0);
}
 
TEST(VarianceTest, NumericalStability)
{
  // Test Welford's algorithm with large values
  vector<double> v = {1e10, 1e10 + 1, 1e10 + 2};
  double var = variance(v);
  EXPECT_NEAR(var, 1.0, 1e-6);
}
 
// =============================================================================
// Standard Deviation Tests
// =============================================================================
 
TEST(StddevTest, Basic)
{
  vector<double> v = {2.0, 4.0, 4.0, 4.0, 5.0, 5.0, 7.0, 9.0};
  double s = stddev(v);
  // Sample stddev of this data is approximately 2.14
  EXPECT_NEAR(s, 2.14, 0.1);
}
 
TEST(StddevTest, IsSquareRootOfVariance)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_NEAR(stddev(v), sqrt(variance(v)), 1e-10);
}
 
// =============================================================================
// Min/Max Tests
// =============================================================================
 
TEST(MinMaxTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(min_value(v), invalid_argument);
  EXPECT_THROW(max_value(v), invalid_argument);
  EXPECT_THROW(min_max(v), invalid_argument);
}
 
TEST(MinMaxTest, SingleElement)
{
  vector<double> v = {5.0};
  EXPECT_DOUBLE_EQ(min_value(v), 5.0);
  EXPECT_DOUBLE_EQ(max_value(v), 5.0);
  auto [min_v, max_v] = min_max(v);
  EXPECT_DOUBLE_EQ(min_v, 5.0);
  EXPECT_DOUBLE_EQ(max_v, 5.0);
}
 
TEST(MinMaxTest, MultipleElements)
{
  vector<double> v = {3.0, 1.0, 4.0, 1.0, 5.0, 9.0, 2.0};
  EXPECT_DOUBLE_EQ(min_value(v), 1.0);
  EXPECT_DOUBLE_EQ(max_value(v), 9.0);
  auto [min_v, max_v] = min_max(v);
  EXPECT_DOUBLE_EQ(min_v, 1.0);
  EXPECT_DOUBLE_EQ(max_v, 9.0);
}
 
TEST(MinMaxTest, NegativeNumbers)
{
  vector<double> v = {-5.0, -1.0, -10.0, -3.0};
  EXPECT_DOUBLE_EQ(min_value(v), -10.0);
  EXPECT_DOUBLE_EQ(max_value(v), -1.0);
}
 
// =============================================================================
// Percentile Tests
// =============================================================================
 
TEST(PercentileTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(percentile(v, 50), invalid_argument);
}
 
TEST(PercentileTest, OutOfRangeThrows)
{
  vector<double> v = {1.0, 2.0, 3.0};
  EXPECT_THROW(percentile(v, -1), invalid_argument);
  EXPECT_THROW(percentile(v, 101), invalid_argument);
}
 
TEST(PercentileTest, Percentile0)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(percentile(v, 0), 1.0);
}
 
TEST(PercentileTest, Percentile100)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(percentile(v, 100), 5.0);
}
 
TEST(PercentileTest, Percentile50IsMedian)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(percentile(v, 50), median(v));
}
 
TEST(PercentileTest, UnsortedInput)
{
  vector<double> v = {5.0, 1.0, 3.0, 2.0, 4.0};
  EXPECT_DOUBLE_EQ(percentile(v, 50), 3.0);
}
 
// =============================================================================
// Median Tests
// =============================================================================
 
TEST(MedianTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(median(v), invalid_argument);
}
 
TEST(MedianTest, SingleElement)
{
  vector<double> v = {5.0};
  EXPECT_DOUBLE_EQ(median(v), 5.0);
}
 
TEST(MedianTest, OddCount)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(median(v), 3.0);
}
 
TEST(MedianTest, EvenCount)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0};
  EXPECT_DOUBLE_EQ(median(v), 2.5);
}
 
TEST(MedianTest, UnsortedData)
{
  vector<double> v = {5.0, 1.0, 3.0};
  EXPECT_DOUBLE_EQ(median(v), 3.0);
}
 
// =============================================================================
// Quartiles and IQR Tests
// =============================================================================
 
TEST(QuartilesTest, Basic)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0};
  auto [q1, q2, q3] = quartiles(v);
  EXPECT_NEAR(q1, 3.25, 0.01);
  EXPECT_NEAR(q2, 5.5, 0.01);
  EXPECT_NEAR(q3, 7.75, 0.01);
}
 
TEST(IqrTest, Basic)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0};
  auto [q1, q2, q3] = quartiles(v);
  EXPECT_NEAR(iqr(v), q3 - q1, 0.01);
}
 
// =============================================================================
// Mode Tests
// =============================================================================
 
TEST(ModeTest, EmptyContainerThrows)
{
  vector<int> v;
  EXPECT_THROW(mode(v), invalid_argument);
}
 
TEST(ModeTest, SingleElement)
{
  vector<int> v = {5};
  EXPECT_EQ(mode(v), 5);
}
 
TEST(ModeTest, AllDifferent)
{
  vector<int> v = {1, 2, 3, 4, 5};
  // When all frequencies are equal, returns first encountered
  int m = mode(v);
  EXPECT_TRUE(m >= 1 and m <= 5);
}
 
TEST(ModeTest, ClearMode)
{
  vector<int> v = {1, 2, 2, 3, 3, 3, 4};
  EXPECT_EQ(mode(v), 3);
}
 
TEST(ModeTest, Multimodal)
{
  vector<int> v = {1, 1, 2, 2, 3};
  EXPECT_TRUE(is_multimodal(v));
}
 
TEST(ModeTest, NotMultimodal)
{
  vector<int> v = {1, 2, 2, 3};
  EXPECT_FALSE(is_multimodal(v));
}
 
// =============================================================================
// Skewness Tests
// =============================================================================
 
TEST(SkewnessTest, TooFewElementsThrows)
{
  vector<double> v = {1.0, 2.0};
  EXPECT_THROW(skewness(v), invalid_argument);
}
 
TEST(SkewnessTest, SymmetricDistribution)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_NEAR(skewness(v), 0.0, 0.1);
}
 
TEST(SkewnessTest, RightSkewed)
{
  vector<double> v = {1.0, 1.0, 1.0, 1.0, 10.0};
  EXPECT_GT(skewness(v), 0.0);  // Positive skewness
}
 
TEST(SkewnessTest, LeftSkewed)
{
  vector<double> v = {10.0, 10.0, 10.0, 10.0, 1.0};
  EXPECT_LT(skewness(v), 0.0);  // Negative skewness
}
 
TEST(SkewnessTest, ConstantValues)
{
  vector<double> v = {5.0, 5.0, 5.0, 5.0};
  EXPECT_DOUBLE_EQ(skewness(v), 0.0);
}
 
// =============================================================================
// Kurtosis Tests
// =============================================================================
 
TEST(KurtosisTest, TooFewElementsThrows)
{
  vector<double> v = {1.0, 2.0, 3.0};
  EXPECT_THROW(kurtosis(v), invalid_argument);
}
 
TEST(KurtosisTest, UniformDistribution)
{
  // Uniform distribution has negative excess kurtosis
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0};
  double k = kurtosis(v);
  EXPECT_LT(k, 0.0);  // Platykurtic
}
 
TEST(KurtosisTest, ConstantValues)
{
  vector<double> v = {5.0, 5.0, 5.0, 5.0, 5.0};
  EXPECT_DOUBLE_EQ(kurtosis(v), 0.0);
}
 
// =============================================================================
// Coefficient of Variation Tests
// =============================================================================
 
TEST(CoefficientOfVariationTest, ZeroMeanThrows)
{
  vector<double> v = {-1.0, 1.0};
  EXPECT_THROW(coefficient_of_variation(v), invalid_argument);
}
 
TEST(CoefficientOfVariationTest, Basic)
{
  vector<double> v = {10.0, 10.0, 10.0, 10.0};
  EXPECT_DOUBLE_EQ(coefficient_of_variation(v), 0.0);
}
 
TEST(CoefficientOfVariationTest, PositiveValue)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  double cv = coefficient_of_variation(v);
  EXPECT_GT(cv, 0.0);
}
 
// =============================================================================
// Covariance Tests
// =============================================================================
 
TEST(CovarianceTest, DifferentSizesThrows)
{
  vector<double> x = {1.0, 2.0, 3.0};
  vector<double> y = {1.0, 2.0};
  EXPECT_THROW(covariance(x, y), invalid_argument);
}
 
TEST(CovarianceTest, TooFewElementsThrows)
{
  vector<double> x = {1.0};
  vector<double> y = {1.0};
  EXPECT_THROW(covariance(x, y, false), invalid_argument);
}
 
TEST(CovarianceTest, PerfectPositive)
{
  vector<double> x = {1.0, 2.0, 3.0, 4.0, 5.0};
  vector<double> y = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_GT(covariance(x, y), 0.0);
}
 
TEST(CovarianceTest, PerfectNegative)
{
  vector<double> x = {1.0, 2.0, 3.0, 4.0, 5.0};
  vector<double> y = {5.0, 4.0, 3.0, 2.0, 1.0};
  EXPECT_LT(covariance(x, y), 0.0);
}
 
TEST(CovarianceTest, NoCorrelation)
{
  vector<double> x = {1.0, 2.0, 3.0};
  vector<double> y = {2.0, 2.0, 2.0};  // Constant
  EXPECT_NEAR(covariance(x, y), 0.0, 1e-10);
}
 
// =============================================================================
// Correlation Tests
// =============================================================================
 
TEST(CorrelationTest, PerfectPositive)
{
  vector<double> x = {1.0, 2.0, 3.0, 4.0, 5.0};
  vector<double> y = {2.0, 4.0, 6.0, 8.0, 10.0};
  EXPECT_NEAR(correlation(x, y), 1.0, 1e-10);
}
 
TEST(CorrelationTest, PerfectNegative)
{
  vector<double> x = {1.0, 2.0, 3.0, 4.0, 5.0};
  vector<double> y = {10.0, 8.0, 6.0, 4.0, 2.0};
  EXPECT_NEAR(correlation(x, y), -1.0, 1e-10);
}
 
TEST(CorrelationTest, ZeroVarianceThrows)
{
  vector<double> x = {1.0, 1.0, 1.0};  // Constant
  vector<double> y = {1.0, 2.0, 3.0};
  EXPECT_THROW(correlation(x, y), invalid_argument);
}
 
TEST(CorrelationTest, PartialCorrelation)
{
  vector<double> x = {1.0, 2.0, 3.0, 4.0, 5.0};
  vector<double> y = {1.0, 2.0, 1.5, 3.5, 5.0};
  double r = correlation(x, y);
  EXPECT_GT(r, 0.0);
  EXPECT_LT(r, 1.0);
}
 
// =============================================================================
// Histogram Tests
// =============================================================================
 
TEST(HistogramTest, ZeroBinsThrows)
{
  vector<double> v = {1.0, 2.0, 3.0};
  EXPECT_THROW(histogram(v, 0), invalid_argument);
}
 
TEST(HistogramTest, EmptyContainerThrows)
{
  vector<double> v;
  EXPECT_THROW(histogram(v, 5), invalid_argument);
}
 
TEST(HistogramTest, SingleBin)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  auto h = histogram(v, 1);
  EXPECT_EQ(h.size(), 1u);
  EXPECT_EQ(h[0].second, 5u);  // All in one bin
}
 
TEST(HistogramTest, MultipleBins)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  auto h = histogram(v, 5);
  EXPECT_EQ(h.size(), 5u);
  // Each value should fall into its own bin
  size_t total = 0;
  for (const auto & [center, count] : h)
    total += count;
  EXPECT_EQ(total, 5u);
}
 
TEST(HistogramTest, ConstantValues)
{
  vector<double> v = {5.0, 5.0, 5.0, 5.0};
  auto h = histogram(v, 3);
  // All values are equal, should be in single bin
  EXPECT_EQ(h.size(), 1u);
  EXPECT_EQ(h[0].second, 4u);
}
 
// =============================================================================
// Stats Structure Tests
// =============================================================================
 
TEST(StatsTest, DefaultConstruction)
{
  Stats<double> s;
  EXPECT_EQ(s.count, 0u);
  EXPECT_FALSE(s.is_valid());
}
 
TEST(StatsTest, RangeMethod)
{
  Stats<double> s;
  s.min = 1.0;
  s.max = 10.0;
  EXPECT_DOUBLE_EQ(s.range(), 9.0);
}
 
// =============================================================================
// compute_all_stats Tests
// =============================================================================
 
TEST(ComputeAllStatsTest, EmptyContainer)
{
  vector<double> v;
  auto s = compute_all_stats(v);
  EXPECT_EQ(s.count, 0u);
  EXPECT_FALSE(s.is_valid());
}
 
TEST(ComputeAllStatsTest, SingleElement)
{
  vector<double> v = {5.0};
  auto s = compute_all_stats(v);
  EXPECT_EQ(s.count, 1u);
  EXPECT_TRUE(s.is_valid());
  EXPECT_DOUBLE_EQ(s.mean, 5.0);
  EXPECT_DOUBLE_EQ(s.sum, 5.0);
  EXPECT_DOUBLE_EQ(s.min, 5.0);
  EXPECT_DOUBLE_EQ(s.max, 5.0);
  EXPECT_DOUBLE_EQ(s.median, 5.0);
}
 
TEST(ComputeAllStatsTest, MultipleElements)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0};
  auto s = compute_all_stats(v);
 
  EXPECT_EQ(s.count, 5u);
  EXPECT_DOUBLE_EQ(s.sum, 15.0);
  EXPECT_DOUBLE_EQ(s.mean, 3.0);
  EXPECT_DOUBLE_EQ(s.min, 1.0);
  EXPECT_DOUBLE_EQ(s.max, 5.0);
  EXPECT_DOUBLE_EQ(s.median, 3.0);
  EXPECT_GT(s.variance, 0.0);
  EXPECT_GT(s.stddev, 0.0);
  EXPECT_DOUBLE_EQ(s.range(), 4.0);
}
 
TEST(ComputeAllStatsTest, QuartilesComputed)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0};
  auto s = compute_all_stats(v);
 
  EXPECT_LT(s.q1, s.median);
  EXPECT_LT(s.median, s.q3);
  EXPECT_NEAR(s.iqr, s.q3 - s.q1, 0.001);
}
 
TEST(ComputeAllStatsTest, HigherMomentsComputed)
{
  vector<double> v = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0};
  auto s = compute_all_stats(v);
 
  // Symmetric distribution should have skewness near 0
  EXPECT_NEAR(s.skewness, 0.0, 0.1);
  // Kurtosis is computed for 10+ elements
  // Uniform has negative excess kurtosis
  EXPECT_LT(s.kurtosis, 0.0);
}
 
// =============================================================================
// Legacy compute_stats Tests (backward compatibility)
// =============================================================================
 
TEST(LegacyComputeStatsTest, ArrayVersion)
{
  double data[] = {5.0, 1.0, 3.0, 2.0, 4.0};
  double avg, var, med, min_val, max_val;
 
  compute_stats(data, 0, 4, avg, var, med, min_val, max_val);
 
  EXPECT_DOUBLE_EQ(avg, 3.0);
  EXPECT_DOUBLE_EQ(med, 3.0);
  EXPECT_DOUBLE_EQ(min_val, 1.0);
  EXPECT_DOUBLE_EQ(max_val, 5.0);
  EXPECT_GT(var, 0.0);
}
 
TEST(LegacyComputeStatsTest, ArrayWithOffset)
{
  double data[] = {100.0, 1.0, 2.0, 3.0, 100.0};
  double avg, var, med, min_val, max_val;
 
  // Only use data[1..3]
  compute_stats(data, 1, 3, avg, var, med, min_val, max_val);
 
  EXPECT_DOUBLE_EQ(avg, 2.0);
  EXPECT_DOUBLE_EQ(med, 2.0);
  EXPECT_DOUBLE_EQ(min_val, 1.0);
  EXPECT_DOUBLE_EQ(max_val, 3.0);
}
 
TEST(LegacyComputeStatsTest, ContainerVersion)
{
  vector<double> v = {5.0, 1.0, 3.0, 2.0, 4.0};
  double avg, var, med, min_val, max_val;
 
  compute_stats(v, avg, var, med, min_val, max_val);
 
  EXPECT_DOUBLE_EQ(avg, 3.0);
  EXPECT_DOUBLE_EQ(med, 3.0);
  EXPECT_DOUBLE_EQ(min_val, 1.0);
  EXPECT_DOUBLE_EQ(max_val, 5.0);
}
 
TEST(LegacyComputeStatsTest, EvenCount)
{
  double data[] = {1.0, 2.0, 3.0, 4.0};
  double avg, var, med, min_val, max_val;
 
  compute_stats(data, 0, 3, avg, var, med, min_val, max_val);
 
  EXPECT_DOUBLE_EQ(med, 2.5);  // (2+3)/2
}
 
TEST(LegacyComputeStatsTest, EmptyRange)
{
  double data[] = {1.0, 2.0, 3.0};
  double avg, var, med, min_val, max_val;
 
  // l > r means empty range
  compute_stats(data, 2, 1, avg, var, med, min_val, max_val);
 
  EXPECT_DOUBLE_EQ(avg, 0.0);
  EXPECT_DOUBLE_EQ(var, 0.0);
  EXPECT_DOUBLE_EQ(med, 0.0);
}
 
// =============================================================================
// Edge Cases and Numerical Stability
// =============================================================================
 
TEST(EdgeCaseTest, VeryLargeNumbers)
{
  vector<double> v = {1e15, 1e15 + 1, 1e15 + 2, 1e15 + 3, 1e15 + 4};
  auto s = compute_all_stats(v);
 
  // Mean should be accurate
  EXPECT_NEAR(s.mean, 1e15 + 2, 1.0);
  // Variance should be stable (Welford's algorithm)
  EXPECT_NEAR(s.variance, 2.5, 0.1);
}
 
TEST(EdgeCaseTest, VerySmallNumbers)
{
  vector<double> v = {1e-15, 2e-15, 3e-15, 4e-15, 5e-15};
  auto s = compute_all_stats(v);
 
  EXPECT_NEAR(s.mean, 3e-15, 1e-16);
  EXPECT_GT(s.variance, 0.0);
}
 
TEST(EdgeCaseTest, MixedSigns)
{
  vector<double> v = {-100.0, -50.0, 0.0, 50.0, 100.0};
  auto s = compute_all_stats(v);
 
  EXPECT_DOUBLE_EQ(s.mean, 0.0);
  EXPECT_DOUBLE_EQ(s.median, 0.0);
}
 
// =============================================================================
// Container Type Tests
// =============================================================================
 
TEST(ContainerTest, WorksWithArray)
{
  array<double, 5> a = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(mean(a), 3.0);
  EXPECT_DOUBLE_EQ(median(a), 3.0);
}
 
TEST(ContainerTest, WorksWithList)
{
  list<double> l = {1.0, 2.0, 3.0, 4.0, 5.0};
  EXPECT_DOUBLE_EQ(mean(l), 3.0);
  EXPECT_DOUBLE_EQ(median(l), 3.0);
}
 
TEST(ContainerTest, WorksWithDynArray)
{
  DynArray<double> a;
  a.append(1.0);
  a.append(2.0);
  a.append(3.0);
  a.append(4.0);
  a.append(5.0);
 
  EXPECT_DOUBLE_EQ(mean(a), 3.0);
  auto s = compute_all_stats(a);
  EXPECT_EQ(s.count, 5u);
}
 
// =============================================================================
// Type Tests
// =============================================================================
 
TEST(TypeTest, WorksWithInt)
{
  vector<int> v = {1, 2, 3, 4, 5};
  EXPECT_EQ(sum(v), 15);
  EXPECT_EQ(mode(v), 1);  // All unique, returns first
}
 
TEST(TypeTest, WorksWithFloat)
{
  vector<float> v = {1.0f, 2.0f, 3.0f, 4.0f, 5.0f};
  EXPECT_FLOAT_EQ(mean(v), 3.0f);
}
 
int main(int argc, char **argv)
{
  ::testing::InitGoogleTest(&argc, argv);
  return RUN_ALL_TESTS();
}