public static void CompareStratMeansVar(string StratModel1, string StratModel2, out List <string> labels, out double[] meanDiff, out double[] varDiff, out double[] meanPvalues, out double[] varPvalues)
{
meanPvalues = null;
varPvalues = null;
meanDiff = null;
varDiff = null;
labels = null;
dataPrepClusterKmean dpc1 = new dataPrepClusterKmean();
dpc1.buildModel(StratModel1);
KMeans km1 = (KMeans)dpc1.Model;
dataPrepClusterKmean dpc2 = new dataPrepClusterKmean();
dpc2.buildModel(StratModel2);
List <string> labels2 = dpc2.Labels;
KMeans km2 = (KMeans)dpc2.Model;
int nPv1 = km1.Clusters.Count;
int nPv2 = km2.Clusters.Count;
if (nPv1 != nPv2)
{
System.Windows.Forms.MessageBox.Show("Not the same number of strata! Models are not comparable!");
return;
}
labels = dpc1.Labels;
meanPvalues = new double[nPv1];
varPvalues = new double[nPv2];
meanDiff = new double[nPv1];
varDiff = new double[nPv2];
foreach (string l in labels)
{
int ind1 = labels.IndexOf(l);
int ind2 = labels2.IndexOf(l);
KMeansCluster kmC1 = km1.Clusters[ind1];
KMeansCluster kmC2 = km2.Clusters[ind2];
double[] means1 = kmC1.Mean;
double[] means2 = kmC2.Mean;
double[,] cov1 = kmC1.Covariance;
double[,] cov2 = kmC2.Covariance;
double[] meanDiffArr = new double[means1.Length];
double[] varDiffArr = new double[means1.Length];
for (int i = 0; i < means1.Length; i++)
{
meanDiffArr[i] = means1[i] - means2[i];
varDiffArr[i] = cov1[i, i] - cov2[i, i];
}
meanDiff[ind1] = meanDiffArr.Average();
varDiff[ind1] = varDiffArr.Average();
double m, v;
PairedTTestPValues(means1, cov1, means2, cov2, out m, out v);
meanPvalues[ind1] = m;
varPvalues[ind1] = v;
}
}
public static void CompareStratMeansVar(string StratModel1, string StratModel2, out List<string> labels, out double[] meanDiff, out double[] varDiff, out double[] meanPvalues, out double[] varPvalues)
{
meanPvalues = null;
varPvalues = null;
meanDiff = null;
varDiff = null;
labels = null;
dataPrepClusterKmean dpc1 = new dataPrepClusterKmean();
dpc1.buildModel(StratModel1);
KMeans km1 = (KMeans)dpc1.Model;
dataPrepClusterKmean dpc2 = new dataPrepClusterKmean();
dpc2.buildModel(StratModel2);
List<string> labels2 = dpc2.Labels;
KMeans km2 = (KMeans)dpc2.Model;
int nPv1 = km1.Clusters.Count;
int nPv2 = km2.Clusters.Count;
if (nPv1 != nPv2)
{
System.Windows.Forms.MessageBox.Show("Not the same number of strata! Models are not comparable!");
return;
}
labels = dpc1.Labels;
meanPvalues = new double[nPv1];
varPvalues = new double[nPv2];
meanDiff = new double[nPv1];
varDiff = new double[nPv2];
foreach (string l in labels)
{
int ind1 = labels.IndexOf(l);
int ind2 = labels2.IndexOf(l);
KMeansCluster kmC1 = km1.Clusters[ind1];
KMeansCluster kmC2 = km2.Clusters[ind2];
double[] means1 = kmC1.Mean;
double[] means2 = kmC2.Mean;
double[,] cov1 = kmC1.Covariance;
double[,] cov2 = kmC2.Covariance;
double[] meanDiffArr = new double[means1.Length];
double[] varDiffArr = new double[means1.Length];
for (int i = 0; i < means1.Length; i++)
{
meanDiffArr[i] = means1[i] - means2[i];
varDiffArr[i] = cov1[i, i] - cov2[i, i];
}
meanDiff[ind1] = meanDiffArr.Average();
varDiff[ind1] = varDiffArr.Average();
double m, v;
PairedTTestPValues(means1, cov1, means2, cov2, out m, out v);
meanPvalues[ind1] = m;
varPvalues[ind1] = v;
}
}
public static double[] clusterProportions(string clusterModelPath)
{
dataPrepClusterKmean cls = new dataPrepClusterKmean();
cls.buildModel(clusterModelPath);
int nClusters = ((Accord.MachineLearning.KMeans)cls.Model).Clusters.Count;
double[] prop = new double[nClusters];
for (int i = 0; i < nClusters; i++)
{
Accord.MachineLearning.KMeansCluster k = ((Accord.MachineLearning.KMeans)cls.Model).Clusters[i];
prop[i] = k.Proportion;
}
return prop;
}
public static double[] clusterProportions(string clusterModelPath)
{
dataPrepClusterKmean cls = new dataPrepClusterKmean();
cls.buildModel(clusterModelPath);
int nClusters = ((Accord.MachineLearning.KMeans)cls.Model).Clusters.Count;
double[] prop = new double[nClusters];
for (int i = 0; i < nClusters; i++)
{
Accord.MachineLearning.KMeansCluster k = ((Accord.MachineLearning.KMeans)cls.Model).Clusters[i];
prop[i] = k.Proportion;
}
return(prop);
}
public static int[] sampleSizeMaxCluster(string clusterModelPath, double proportionOfMean = 0.1, double alpha = 0.05)
{
dataPrepClusterKmean cls = new dataPrepClusterKmean();
cls.buildModel(clusterModelPath);
int nClusters = ((Accord.MachineLearning.KMeans)cls.Model).Clusters.Count;
int[] maxN = new int[nClusters];
for (int i = 0; i < nClusters; i++)
{
Accord.MachineLearning.KMeansCluster k = ((Accord.MachineLearning.KMeans)cls.Model).Clusters[i];
int mx = sampleSizeMaxMean(k.Covariance, k.Mean, proportionOfMean, alpha)[0];
maxN[i] = mx;
}
return(maxN);
}
private static void fillCluserReport(string modelPath, Forms.RunningProcess.frmRunningProcessDialog rp, double proportion = 0.1, double alpha = 0.05)
{
dataPrepClusterKmean clus = new dataPrepClusterKmean();
clus.buildModel(modelPath);
List <string> lbl = clus.Labels;
rp.addMessage("Samples by class (Cluster; number of samples)");
rp.addMessage("-".PadRight(45, '-'));
int[] samples = sampleSizeMaxCluster(modelPath, proportion, alpha);
for (int i = 0; i < samples.Length; i++)
{
rp.addMessage("\t" + lbl[i] + "; " + samples[i].ToString());
}
rp.addMessage("-".PadRight(45, '-'));
rp.addMessage("Total number of samples = " + samples.Sum().ToString());
}
private void buildModel()
{
if (!checkTables())
{
//Console.WriteLine("CheckTables = false");
return;
}
if (!getSampleRatios())
{
//Console.WriteLine("Sample Ratios = false");
return;
}
pca = new dataPrepPrincipleComponents(Sample1, Variables);
cluster = new dataPrepClusterKmean(Sample1, Variables, numberOfBins);
foreach (string s in cntDic.Keys)
{
buildSamples(s);
double[] pValueArr = new double[Variables.Length];
double[] sValueArr = new double[Variables.Length];
//double[] s1Arr = sample1[0];
//double[] s2Arr = sample2[0];
//TwoSampleKolmogorovSmirnovTest test = new TwoSampleKolmogorovSmirnovTest(s1Arr, s2Arr, TwoSampleKolmogorovSmirnovTestHypothesis.SamplesDistributionsAreUnequal);
//getCdfProp(s, 0, test);
//pValue = test.PValue;
//sValue = test.Statistic;
for (int i = 0; i < Variables.Length; i++)
{
double[] s1Arr = sample1[i];
double[] s2Arr = sample2[i];
TwoSampleKolmogorovSmirnovTest test = new TwoSampleKolmogorovSmirnovTest(s1Arr, s2Arr, TwoSampleKolmogorovSmirnovTestHypothesis.SamplesDistributionsAreUnequal);
//Console.WriteLine(test.Significant.ToString());
getCdfProp(s, i, test);
double pValueS = test.PValue;
double sValueS = test.Statistic;
pValueArr[i] = pValueS;
sValueArr[i] = sValueS;
}
pDic.Add(s, pValueArr);
sDic.Add(s, sValueArr);
}
}
private void buildModel(string mdlPath)
{
using (System.IO.StreamReader sr = new System.IO.StreamReader(mdlPath))
{
dataPrepBase.modelTypes mType = (dataPrepBase.modelTypes)Enum.Parse(typeof(dataPrepBase.modelTypes), sr.ReadLine());
if (mType != dataPrepBase.modelTypes.KS)
{
System.Windows.Forms.MessageBox.Show("Not a KS Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
Variables = sr.ReadLine().Split(new char[] { ',' });
StrataField = sr.ReadLine();
Oridinate = System.Convert.ToBoolean(sr.ReadLine());
string[] lbl = sr.ReadLine().Split(new char[] { ',' });
string[] pArr = sr.ReadLine().Split(new char[] { ',' });
string[] sArr = sr.ReadLine().Split(new char[] { ',' });
string[] cArr1 = sr.ReadLine().Split(new char[] { ',' });
string[] cArr2 = sr.ReadLine().Split(new char[] { ',' });
for (int i = 0; i < lbl.Length; i++)
{
string l = lbl[i];
double[] p = (from string str in pArr[i].Split(new char[] { ';' }) select System.Convert.ToDouble(str)).ToArray();
double[] s = (from string str in sArr[i].Split(new char[] { ';' }) select System.Convert.ToDouble(str)).ToArray();
int c1 = System.Convert.ToInt32(cArr1[i]);
int c2 = System.Convert.ToInt32(cArr2[i]);
pDic.Add(l, p);
sDic.Add(l, s);
cntDic.Add(l, new int[] { c1, c2 });
}
double[][] minmax1, minmax2, bp1, bp2;
for (int i = 0; i < lbl.Length; i++)
{
string l = lbl[i];
double[] min1 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] max1 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] min2 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] max2 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
minmax1 = new double[2][];
minmax1[0] = min1;
minmax1[1] = max1;
minMaxDic1.Add(l, minmax1);
minmax2 = new double[2][];
minmax2[0] = min2;
minmax2[1] = max2;
minMaxDic2.Add(l, minmax2);
bp1 = new double[Variables.Length][];
bp2 = new double[Variables.Length][];
for (int j = 0; j < Variables.Length; j++)
{
bp1[j] = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
bp2[j] = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
}
binPropDic1.Add(l, bp1);
binPropDic2.Add(l, bp2);
int[] clusCnt = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToInt32(s)).ToArray();
clusCountDic.Add(l, clusCnt);
int[] clusSampCnt = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToInt32(s)).ToArray();
clusSampleCountDic.Add(l, clusSampCnt);
}
pca = new dataPrepPrincipleComponents();
string pcPath = System.IO.Path.GetDirectoryName(mdlPath) + "\\" + System.IO.Path.GetFileNameWithoutExtension(mdlPath) + "_pca.mdl";
pca.buildModel(pcPath);
cluster = new dataPrepClusterKmean();
string clusterPath = System.IO.Path.GetDirectoryName(mdlPath) + "\\" + System.IO.Path.GetFileNameWithoutExtension(mdlPath) + "_cluster.mdl";
cluster.buildModel(clusterPath);
numberOfBins = cluster.Classes;
sr.Close();
}
}
private void createClusterModel(string[] paramArr)
{
Statistics.dataPrepClusterBase dpC = null;
esriUtil.Statistics.clusterType cType = Statistics.clusterType.KMEANS;
if(paramArr.Length<6)
{
IRaster rs = rsUtil.createRaster(getRaster(paramArr[1]));
int nCls = System.Convert.ToInt32(paramArr[2]);
cType = (esriUtil.Statistics.clusterType)Enum.Parse(typeof(esriUtil.Statistics.clusterType), paramArr[3]);
switch (cType)
{
case esriUtil.Statistics.clusterType.KMEANS:
dpC = (Statistics.dataPrepClusterBase)(new Statistics.dataPrepClusterKmean(rs, nCls));
break;
case esriUtil.Statistics.clusterType.BINARY:
dpC = (Statistics.dataPrepClusterBase)(new Statistics.dataPrepClusterBinary(rs, nCls));
break;
case esriUtil.Statistics.clusterType.GAUSSIANMIXTURE:
dpC = (Statistics.dataPrepClusterBase)(new Statistics.dataPrepClusterGaussian(rs, nCls));
break;
default:
break;
}
}
else
{
ITable tbl = getTable(paramArr[1]);
string[] flds = paramArr[2].Split(new char[]{','});
int nCls = System.Convert.ToInt32(paramArr[3]);
cType = (esriUtil.Statistics.clusterType)Enum.Parse(typeof(esriUtil.Statistics.clusterType), paramArr[4]);
dpC = new Statistics.dataPrepClusterKmean(tbl,flds,nCls);
}
dpC.writeModel(paramArr[paramArr.Length - 1]);
}
private static void fillCluserReport(string modelPath, Forms.RunningProcess.frmRunningProcessDialog rp, double proportion = 0.1, double alpha = 0.05)
{
dataPrepClusterKmean clus = new dataPrepClusterKmean();
clus.buildModel(modelPath);
List<string> lbl = clus.Labels;
rp.addMessage("Samples by class (Cluster; number of samples)");
rp.addMessage("-".PadRight(45, '-'));
int[] samples = sampleSizeMaxCluster(modelPath, proportion, alpha);
for (int i = 0; i < samples.Length; i++)
{
rp.addMessage("\t"+lbl[i] + "; " + samples[i].ToString());
}
rp.addMessage("-".PadRight(45, '-'));
rp.addMessage("Total number of samples = " + samples.Sum().ToString());
}
public static int[] sampleSizeMaxCluster(string clusterModelPath, double proportionOfMean = 0.1, double alpha = 0.05)
{
dataPrepClusterKmean cls = new dataPrepClusterKmean();
cls.buildModel(clusterModelPath);
int nClusters = ((Accord.MachineLearning.KMeans)cls.Model).Clusters.Count;
int[] maxN = new int[nClusters];
for (int i = 0; i < nClusters; i++)
{
Accord.MachineLearning.KMeansCluster k = ((Accord.MachineLearning.KMeans)cls.Model).Clusters[i];
int mx = sampleSizeMaxMean(k.Covariance, k.Mean, proportionOfMean, alpha)[0];
maxN[i] = mx;
}
return maxN;
}
/// <summary>
/// creates a new field called sample and populates yes or no depending on whether that feature should be sampled based on a previously ran cluster analysis
/// </summary>
/// <param name="inputTable"></param>
/// <param name="clusterModelPath"></param>
/// <param name="proportionOfMean"></param>
/// <param name="alpha"></param>
public void selectClusterFeaturesToSample(ITable inputTable, string clusterModelPath, string clusterFieldName="Cluster", double proportionOfMean=0.1, double alpha=0.05, bool weightsEqual=false)
{
IObjectClassInfo2 objInfo2 = (IObjectClassInfo2)inputTable;
if (!objInfo2.CanBypassEditSession())
{
System.Windows.Forms.MessageBox.Show("Input Table participates in a composite relationship. Please export this table as a new table and try again!");
return;
}
esriUtil.Statistics.dataPrepClusterKmean dpC = new Statistics.dataPrepClusterKmean();
dpC.buildModel(clusterModelPath);
List<string> labels = dpC.Labels;
HashSet<string> unqVls = geoUtil.getUniqueValues(inputTable, clusterFieldName);
System.Random rd = new Random();
int[] samplesPerCluster = esriUtil.Statistics.dataPrepSampleSize.sampleSizeMaxCluster(clusterModelPath, proportionOfMean, alpha);
double[] propPerCluster = esriUtil.Statistics.dataPrepSampleSize.clusterProportions(clusterModelPath);
double[] weightsPerCluster = new double[propPerCluster.Length];
double sSamp = System.Convert.ToDouble(samplesPerCluster.Sum());
for (int i = 0; i < weightsPerCluster.Length; i++)
{
weightsPerCluster[i] = propPerCluster[i] / (samplesPerCluster[i] / sSamp);
}
if (weightsEqual)
{
double minProp = weightsPerCluster.Min();
for (int i = 0; i < samplesPerCluster.Length; i++)
{
samplesPerCluster[i] = System.Convert.ToInt32(samplesPerCluster[i] * (weightsPerCluster[i] / minProp));
weightsPerCluster[i] = 1;
}
}
int[] tsPerCluster = new int[propPerCluster.Length];
double[] randomRatioPerClust = new double[propPerCluster.Length];
if (samplesPerCluster.Length != unqVls.Count)
{
System.Windows.Forms.MessageBox.Show("Unique Values in cluster field do not match the number of cluster models!");
return;
}
string sampleFldName = geoUtil.createField(inputTable, "sample", esriFieldType.esriFieldTypeSmallInteger,false);
string weightFldName = geoUtil.createField(inputTable, "weight", esriFieldType.esriFieldTypeDouble,false);
IQueryFilter qf0 = new QueryFilterClass();
qf0.SubFields = clusterFieldName;
string h = "";
IField fld = inputTable.Fields.get_Field(inputTable.FindField(clusterFieldName));
if (fld.Type == esriFieldType.esriFieldTypeString) h = "'";
for (int i = 0; i < samplesPerCluster.Length; i++)
{
qf0.WhereClause = clusterFieldName + " = " + h+labels[i]+h;
int tCnt = inputTable.RowCount(qf0);
tsPerCluster[i] = tCnt;
randomRatioPerClust[i] = System.Convert.ToDouble(samplesPerCluster[i]) / tCnt;
}
IQueryFilter qf = new QueryFilterClass();
qf.SubFields = clusterFieldName + "," + sampleFldName + "," + weightFldName;
IWorkspace wks = ((IDataset)inputTable).Workspace;
IWorkspaceEdit wksE = (IWorkspaceEdit)wks;
if (wksE.IsBeingEdited())
{
wksE.StopEditing(true);
}
try
{
ICursor cur = inputTable.Update(qf, false);
int sIndex = cur.FindField(sampleFldName);
int cIndex = cur.FindField(clusterFieldName);
int wIndex = cur.FindField(weightFldName);
IRow rw = cur.NextRow();
while (rw != null)
{
string clustStr = rw.get_Value(cIndex).ToString();
int clust = labels.IndexOf(clustStr);
double w = weightsPerCluster[clust];
double rNum = rd.NextDouble();
int ss = 0;
double r = randomRatioPerClust[clust];
if (rNum < r)
{
ss = 1;
}
rw.set_Value(sIndex, ss);
rw.set_Value(wIndex, w);
cur.UpdateRow(rw);
rw = cur.NextRow();
}
System.Runtime.InteropServices.Marshal.ReleaseComObject(cur);
}
catch(Exception e)
{
System.Windows.Forms.MessageBox.Show(e.ToString());
}
}
private void buildModel(string mdlPath)
{
using (System.IO.StreamReader sr = new System.IO.StreamReader(mdlPath))
{
dataPrepBase.modelTypes mType = (dataPrepBase.modelTypes)Enum.Parse(typeof(dataPrepBase.modelTypes), sr.ReadLine());
if (mType != dataPrepBase.modelTypes.KS)
{
System.Windows.Forms.MessageBox.Show("Not a KS Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
Variables = sr.ReadLine().Split(new char[] { ',' });
StrataField = sr.ReadLine();
Oridinate = System.Convert.ToBoolean(sr.ReadLine());
string[] lbl = sr.ReadLine().Split(new char[] { ',' });
string[] pArr = sr.ReadLine().Split(new char[] { ',' });
string[] sArr = sr.ReadLine().Split(new char[] {','});
string[] cArr1 = sr.ReadLine().Split(new char[] { ',' });
string[] cArr2 = sr.ReadLine().Split(new char[] { ',' });
for (int i = 0; i < lbl.Length; i++)
{
string l = lbl[i];
double[] p = (from string str in pArr[i].Split(new char[]{';'}) select System.Convert.ToDouble(str)).ToArray();
double[] s = (from string str in sArr[i].Split(new char[] { ';' }) select System.Convert.ToDouble(str)).ToArray();
int c1 = System.Convert.ToInt32(cArr1[i]);
int c2 = System.Convert.ToInt32(cArr2[i]);
pDic.Add(l, p);
sDic.Add(l, s);
cntDic.Add(l, new int[]{c1,c2});
}
double[][] minmax1, minmax2, bp1, bp2;
for (int i = 0; i < lbl.Length; i++)
{
string l = lbl[i];
double[] min1 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] max1 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] min2 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
double[] max2 = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
minmax1 = new double[2][];
minmax1[0] = min1;
minmax1[1] = max1;
minMaxDic1.Add(l, minmax1);
minmax2 = new double[2][];
minmax2[0] = min2;
minmax2[1] = max2;
minMaxDic2.Add(l,minmax2);
bp1 = new double[Variables.Length][];
bp2 = new double[Variables.Length][];
for (int j = 0; j < Variables.Length; j++)
{
bp1[j] = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
bp2[j] = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToDouble(s)).ToArray();
}
binPropDic1.Add(l, bp1);
binPropDic2.Add(l, bp2);
int[] clusCnt = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToInt32(s)).ToArray();
clusCountDic.Add(l, clusCnt);
int[] clusSampCnt = (from string s in sr.ReadLine().Split(new char[] { ',' }) select System.Convert.ToInt32(s)).ToArray();
clusSampleCountDic.Add(l, clusSampCnt);
}
pca = new dataPrepPrincipleComponents();
string pcPath = System.IO.Path.GetDirectoryName(mdlPath) + "\\" + System.IO.Path.GetFileNameWithoutExtension(mdlPath) + "_pca.mdl";
pca.buildModel(pcPath);
cluster = new dataPrepClusterKmean();
string clusterPath = System.IO.Path.GetDirectoryName(mdlPath) + "\\" + System.IO.Path.GetFileNameWithoutExtension(mdlPath) + "_cluster.mdl";
cluster.buildModel(clusterPath);
numberOfBins = cluster.Classes;
sr.Close();
}
}
private void buildModel()
{
if (!checkTables())
{
//Console.WriteLine("CheckTables = false");
return;
}
if (!getSampleRatios())
{
//Console.WriteLine("Sample Ratios = false");
return;
}
pca = new dataPrepPrincipleComponents(Sample1, Variables);
cluster = new dataPrepClusterKmean(Sample1, Variables, numberOfBins);
foreach(string s in cntDic.Keys)
{
buildSamples(s);
double[] pValueArr = new double[Variables.Length];
double[] sValueArr = new double[Variables.Length];
//double[] s1Arr = sample1[0];
//double[] s2Arr = sample2[0];
//TwoSampleKolmogorovSmirnovTest test = new TwoSampleKolmogorovSmirnovTest(s1Arr, s2Arr, TwoSampleKolmogorovSmirnovTestHypothesis.SamplesDistributionsAreUnequal);
//getCdfProp(s, 0, test);
//pValue = test.PValue;
//sValue = test.Statistic;
for (int i = 0; i < Variables.Length; i++)
{
double[] s1Arr = sample1[i];
double[] s2Arr = sample2[i];
TwoSampleKolmogorovSmirnovTest test = new TwoSampleKolmogorovSmirnovTest(s1Arr, s2Arr, TwoSampleKolmogorovSmirnovTestHypothesis.SamplesDistributionsAreUnequal);
//Console.WriteLine(test.Significant.ToString());
getCdfProp(s, i, test);
double pValueS = test.PValue;
double sValueS = test.Statistic;
pValueArr[i] = pValueS;
sValueArr[i] = sValueS;
}
pDic.Add(s, pValueArr);
sDic.Add(s, sValueArr);
}
}