public DataTable OutputCorrelation()
{
NumberStringChecker nsc = new NumberStringChecker();
List <List <double> > listOfLists = new List <List <double> >();
DataTable correlationDataTable = new DataTable("Correlation");
for (int j = 0; j < dataTable.Columns.Count; j++)
{
List <double> list = new List <double>();
//verifica daca elem initial al coloanei e numeric
if (!nsc.IsString(dataTable.Rows[0].ItemArray[j].ToString()))
{
for (int i = 0; i < dataTable.Rows.Count; i++)
{
String s = dataTable.Rows[i].ItemArray[j].ToString();
Double d = 0;
//adauga in lista doar daca celula poate fi convertita la double
if (Double.TryParse(s, out d))
{
list.Add(d);
}
}
listOfLists.Add(list);
DataColumn nameDataColumn = new DataColumn(dataTable.Columns[j].ColumnName.ToString());
correlationDataTable.Columns.Add(nameDataColumn);
}
}
double[][] matrix = new double[listOfLists.Count][];
for (int increment = 0; increment < listOfLists.Count; increment++)
{
matrix[increment] = listOfLists[increment].ToArray();
}
listOfLists.Clear();
//verific daca matricea e simetrica
Boolean isSymmetricMatrix = true;
int numberElementsFirstRow = matrix[0].ToArray().Length;
for (int increment = 0; increment < matrix.Length && isSymmetricMatrix == true; increment++)
{
if (matrix[increment].Length != numberElementsFirstRow)
{
isSymmetricMatrix = false;
}
}
if (isSymmetricMatrix)
{
var corr = MathNet.Numerics.Statistics.Correlation.SpearmanMatrix(matrix);
for (int increment = 0; increment < corr.ColumnCount; increment++)
{
DataRow dataRow = correlationDataTable.NewRow();
for (int increment2 = 0; increment2 < corr.RowCount; increment2++)
{
dataRow[increment2] = corr[increment2, increment];
}
correlationDataTable.Rows.Add(dataRow);
}
return(correlationDataTable);
}
else
{
MessageBox.Show("Correlation data is invalid.");
return(null);
}
}
public void Extrapolation(DataGridView dataGridViewStatistics, int selectedColumn, int nrOfPredictions)
{
NumberStringChecker nsc = new NumberStringChecker();
List <double> listX = new List <double>();
List <double> listY = new List <double>();
if (!nsc.IsString(dataGridView.Rows[0].Cells[selectedColumn].Value.ToString()))
{
for (int i = 0; i < dataGridView.Rows.Count; i++)
{
String s = dataGridView.Rows[i].Cells[selectedColumn].Value.ToString();
Double d = 0;
if (Double.TryParse(s, out d))
{
listY.Add(d);
}
}
}
double[] xdata = new double[dataGridView.Rows.Count];
for (int i = 0; i < xdata.Length; i++)
{
xdata[i] = i + 1;
}
listX.Clear();
double[] ydata = listY.ToArray();
listY.Clear();
if (xdata.Length == ydata.Length)
{
Tuple <double, double> t = MathNet.Numerics.LinearRegression.SimpleRegression.Fit(xdata, ydata);
double a = t.Item1;
double b = t.Item2;
Console.WriteLine("y = {0} + {1} * x", a, b);
DataTable extrapolationDataTable = new DataTable("Extrapolation");
DataColumn numberDataColumn = new DataColumn("Number");
DataColumn valueDataColumn = new DataColumn("Value");
DataColumn extrapolatedDataColumn = new DataColumn("Extrapolated");
extrapolationDataTable.Columns.Add(numberDataColumn);
extrapolationDataTable.Columns.Add(valueDataColumn);
extrapolationDataTable.Columns.Add(extrapolatedDataColumn);
for (int i = 0; i < xdata.Length + nrOfPredictions; i++)
{
if (i < xdata.Length)
{
DataRow dataRow = extrapolationDataTable.NewRow();
dataRow["Number"] = xdata[i];
dataRow["Value"] = ydata[i];
extrapolationDataTable.Rows.Add(dataRow);
}
else
{
DataRow dataRow = extrapolationDataTable.NewRow();
dataRow["Number"] = i + 1;
dataRow["Value"] = a + b * (i + 1);
dataRow["Extrapolated"] = "Yes";
extrapolationDataTable.Rows.Add(dataRow);
}
}
dataGridViewStatistics.DataSource = null;
dataGridViewStatistics.DataSource = extrapolationDataTable;
}
else
{
MessageBox.Show("Extrapolation failed.");
}
}
public DataTable OutputDescriptiveStatisticsPopulation()
{
NumberStringChecker nsc = new NumberStringChecker();
DataTable descriptiveStatisticsDataTable = new DataTable("Descriptive Statistics");
DataColumn nameDataColumn = new DataColumn("Name");
DataColumn sizeDataColumn = new DataColumn("Size");
DataColumn maximumDataColumn = new DataColumn("Maximum");
DataColumn minimumDataColumn = new DataColumn("Minimum");
DataColumn meanDataColumn = new DataColumn("Mean");
DataColumn medianDataColumn = new DataColumn("Median");
DataColumn varianceDataColumn = new DataColumn("Variance");
DataColumn standardDeviationDataColumn = new DataColumn("StandardDeviation");
DataColumn kurtosisDataColumn = new DataColumn("Kurtosis");
DataColumn skewnessDataColumn = new DataColumn("Skewness");
descriptiveStatisticsDataTable.Columns.Add(nameDataColumn);
descriptiveStatisticsDataTable.Columns.Add(sizeDataColumn);
descriptiveStatisticsDataTable.Columns.Add(maximumDataColumn);
descriptiveStatisticsDataTable.Columns.Add(minimumDataColumn);
descriptiveStatisticsDataTable.Columns.Add(meanDataColumn);
descriptiveStatisticsDataTable.Columns.Add(medianDataColumn);
descriptiveStatisticsDataTable.Columns.Add(varianceDataColumn);
descriptiveStatisticsDataTable.Columns.Add(standardDeviationDataColumn);
descriptiveStatisticsDataTable.Columns.Add(kurtosisDataColumn);
descriptiveStatisticsDataTable.Columns.Add(skewnessDataColumn);
for (int j = 0; j < dataTable.Columns.Count; j++)
{
List <double> list = new List <double>();
//verifica daca elem initial al coloanei e numeric
if (!nsc.IsString(dataTable.Rows[0].ItemArray[j].ToString()))
{
for (int i = 0; i < dataTable.Rows.Count; i++)
{
String s = dataTable.Rows[i].ItemArray[j].ToString();
Double d = 0;
//adauga in lista doar daca celula poate fi convertita la double
if (Double.TryParse(s, out d))
{
list.Add(d);
}
}
DataRow dataRow = descriptiveStatisticsDataTable.NewRow();
dataRow["Name"] = dataTable.Columns[j].ColumnName.ToString();
dataRow["Size"] = list.Count();
double maximum = Statistics.Maximum(list);
double minimum = Statistics.Minimum(list);
double mean = Statistics.Mean(list);
double median = Statistics.Median(list);
double variance = Statistics.PopulationVariance(list);
double standardDeviation = Statistics.PopulationStandardDeviation(list);
double kurtosis = Statistics.PopulationKurtosis(list);
double skewness = Statistics.PopulationSkewness(list);
dataRow["Maximum"] = maximum;
dataRow["Minimum"] = minimum;
dataRow["Mean"] = mean;
dataRow["Median"] = median;
dataRow["Variance"] = variance;
dataRow["StandardDeviation"] = standardDeviation;
dataRow["Kurtosis"] = kurtosis;
dataRow["Skewness"] = skewness;
descriptiveStatisticsDataTable.Rows.Add(dataRow);
list.Clear();
}
}
return(descriptiveStatisticsDataTable);
}
public void LinearRegressionSimple(DataGridView dataGridViewStatistics, int[] columns)
{
NumberStringChecker nsc = new NumberStringChecker();
List <double> listX = new List <double>();
List <double> listY = new List <double>();
for (int j = 0; j < columns.Length && columns.Length == 2 && dataGridView.Columns.Count >= 2; j++)
{
int selectedColumn = columns[j];
if (!nsc.IsString(dataGridView.Rows[0].Cells[selectedColumn].Value.ToString()))
{
for (int i = 0; i < dataGridView.Rows.Count; i++)
{
String s = dataGridView.Rows[i].Cells[selectedColumn].Value.ToString();
Double d = 0;
if (Double.TryParse(s, out d))
{
if (selectedColumn == columns[0])
{
listX.Add(d);
}
else if (selectedColumn == columns[1])
{
listY.Add(d);
}
}
}
}
}
double[] xdata = listX.ToArray();
listX.Clear();
double[] ydata = listY.ToArray();
listY.Clear();
if (xdata.Length == ydata.Length)
{
Tuple <double, double> t = MathNet.Numerics.LinearRegression.SimpleRegression.Fit(xdata, ydata);
double a = t.Item1;
double b = t.Item2;
Console.WriteLine("y = {0} + {1} * x", a, b);
DataTable linearRegressionDataTable = new DataTable("Linear regression");
DataColumn nameDataColumn = new DataColumn("Name");
DataColumn sizeDataColumn = new DataColumn("Size");
DataColumn interceptDataColumn = new DataColumn("Intercept");
DataColumn slopeDataColumn = new DataColumn("Slope");
linearRegressionDataTable.Columns.Add(nameDataColumn);
linearRegressionDataTable.Columns.Add(sizeDataColumn);
linearRegressionDataTable.Columns.Add(interceptDataColumn);
linearRegressionDataTable.Columns.Add(slopeDataColumn);
DataRow dataRow = linearRegressionDataTable.NewRow();
dataRow["Name"] = dataGridView.Columns[columns[0]].Name.ToString() + " " + dataGridView.Columns[columns[1]].Name.ToString();
dataRow["Size"] = xdata.Length;
dataRow["Intercept"] = a;
dataRow["Slope"] = b;
linearRegressionDataTable.Rows.Add(dataRow);
dataGridViewStatistics.DataSource = null;
dataGridViewStatistics.DataSource = linearRegressionDataTable;
}
else
{
MessageBox.Show("Simple regression failed.");
}
}