private List <double> pearson()
{
PearsonCorrelation cor = new PearsonCorrelation();
List <double> rets = new List <double>();
int totElem = values.Count - 12;
double[] orig = new double[totElem];
double[] sfas;
for (int i = 0; i < orig.Length; i++)
{
orig[i] = values[i + 12];
}
for (int gap = 1; gap < 13; gap++)
{
sfas = new double[totElem];
for (int i = 0; i < sfas.Length; i++)
{
sfas[i] = values[i + 12 - gap];
}
double sim = cor.GetSimilarityScore(orig, sfas);
rets.Add(sim);
}
return(rets);
}
public void PearsonCorrelationTest( )
{
PearsonCorrelation sim = new PearsonCorrelation( );
Assert.Throws <ArgumentException>(() => sim.GetSimilarityScore(p0, q4));
double result = sim.GetSimilarityScore(p0, q0);
Assert.AreEqual(result, -1);
result = sim.GetSimilarityScore(p1, q1);
Assert.AreEqual(result, 1);
result = sim.GetSimilarityScore(p2, q2);
Assert.AreEqual(result, 0);
result = sim.GetSimilarityScore(p3, q3);
Assert.AreEqual(result, 0);
result = sim.GetSimilarityScore(p4, q4);
Assert.AreApproximatelyEqual(result, 0.396059, 0.00001);
result = sim.GetSimilarityScore(p5, q5);
Assert.AreApproximatelyEqual(result, 0.85470, 0.00001);
}
/// <summary>
/// Compute Pearson Correlation between each data column of the inputs and the identified data column of the output
/// </summary>
/// <param name="outputColumn">Indicates the column index of output data </param>
/// <returns></returns>
public double[] GetPearsonCorrelations(int outputColumn)
{
double[] result = new double[InputVectorLength];
PearsonCorrelation pc = new PearsonCorrelation();
for (int i = 0; i < InputVectorLength; i++)
{
result[i] = pc.GetSimilarityScore(this[i, true], this[outputColumn, false]);
}
return(result);
}