public KPCAResult(DataTable inputData, double[,] inputDataDeviationScores, double[,] inputDataStandardScores,
string[] columnNames, AnalysisMethod method, DescriptiveMeasureCollection measures,
double[,] scatterPlotValues, double[,] componentMatrix, PrincipalComponentCollection principalComponents)
{
this.inputData = inputData;
this.inputDataDeviationScores = inputDataDeviationScores;
this.inputDataStandardScores = inputDataStandardScores;
this.columnNames = columnNames;
this.method = method;
this.measures = measures;
this.scatterPlotValues = scatterPlotValues;
this.componentMatrix = componentMatrix;
this.principalComponents = principalComponents;
}
public KPCAResult(DataTable inputData, double[,] inputDataDeviationScores, double[,] inputDataStandardScores,
string[] columnNames, AnalysisMethod method, DescriptiveMeasureCollection measures,
double[,] scatterPlotValues, double[,] componentMatrix, PrincipalComponentCollection principalComponents)
{
this.inputData = inputData;
this.inputDataDeviationScores = inputDataDeviationScores;
this.inputDataStandardScores = inputDataStandardScores;
this.columnNames = columnNames;
this.method = method;
this.measures = measures;
this.scatterPlotValues = scatterPlotValues;
this.componentMatrix = componentMatrix;
this.principalComponents = principalComponents;
}
public string ProcessLinearRegression(DataTable sourceTable, string _dependentName, string _independentName)
{
try
{
double[][] inputs;
double[] outputs;
LinearRegressionOutput output = new LinearRegressionOutput();
// Gets the column of the dependent variable
String dependentName = _dependentName;
DataTable dependent = sourceTable.DefaultView.ToTable(false, dependentName);
String[] independentNames = _independentName.Split(',');
//String[] independentNames = names.ToArray();
DataTable independent = sourceTable.DefaultView.ToTable(false, independentNames);
// Creates the input and output matrices from the source data table
inputs = independent.ToJagged();
outputs = dependent.Columns[dependentName].ToArray();
// Creates the Simple Descriptive Analysis of the given source
var sda = new DescriptiveAnalysis()
{
ColumnNames = independentNames
}.Learn(inputs);
// TODO: Standardize the InputNames/OutputNames properties
// Populates statistics overview tab with analysis data
DescriptiveMeasureCollection DMC = sda.Measures;
DistributionMesure dm = new DistributionMesure()
{
//Column = sda.Measures.GetEnumerator(x => x),
};
//DMC.s
//output.DistributionMeasuresDataSource = dm;
// Creates the Logistic Regression Analysis of the given source
this.lra = new LogisticRegressionAnalysis()
{
Inputs = independentNames,
Output = dependentName
};
// Create the Multiple Linear Regression Analysis of the given source
this.mlr = new MultipleLinearRegressionAnalysis(intercept: true)
{
Inputs = independentNames,
Output = dependentName
};
// Compute the Linear Regression Analysis
MultipleLinearRegression reg = mlr.Learn(inputs, outputs);
LinearRegressionCoefficientCollection LCC = mlr.Coefficients;
List <LinearCoefficients> lcs = new List <LinearCoefficients>();
foreach (var rc in LCC)
{
LinearCoefficients lc = new LinearCoefficients()
{
Name = rc.Name,
Value = rc.Value,
StandardError = rc.StandardError,
TStatistic = rc.TTest.Statistic,
P_ValueofT = rc.TTest.PValue,
FStatistic = rc.FTest.Statistic,
P_ValueofF = rc.FTest.PValue,
ConfidenceUpper = rc.ConfidenceUpper,
ConfidenceLower = rc.ConfidenceLower
};
lcs.Add(lc);
}
output.LinearCoefficientsDataSource = lcs;
//AnovaSourceCollection RDS = mlr.Table;
List <RegressionAnova> acs = new List <RegressionAnova>();
foreach (var RDS in mlr.Table)
{
RegressionAnova ra = new RegressionAnova()
{
Source = RDS.Source,
DegreesOfFreedom = RDS.DegreesOfFreedom,
SumOfSquares = RDS.SumOfSquares,
MeanSquares = RDS.MeanSquares,
FStatistic = RDS.Statistic,
PValueSignificance = (RDS.Significance == null) ? 0 : RDS.Significance.PValue
};
acs.Add(ra);
}
output.RegressionAnovaDataSource = acs;
output.RSquared = mlr.RSquared.ToString("N5");
output.RSquaredAdj = mlr.RSquareAdjusted.ToString("N5");
output.ChiPValue = mlr.ChiSquareTest.PValue.ToString("N5");
output.FPValue = mlr.FTest.PValue.ToString("N5");
output.ZPValue = mlr.ZTest.PValue.ToString("N5");
output.ChiStatistic = mlr.ChiSquareTest.Statistic.ToString("N5");
output.FStatistic = mlr.FTest.Statistic.ToString("N5");
output.ZStatistic = mlr.ZTest.Statistic.ToString("N5");
output.ChiSignificantChecked = mlr.ChiSquareTest.Significant;
output.FSignificantChecked = mlr.FTest.Significant;
output.ZSignificantChecked = mlr.ZTest.Significant;
// Populate projection source table
string[] cols = independentNames;
if (!independentNames.Contains(dependentName))
{
cols = independentNames.Concatenate(dependentName);
}
DataTable projSource = sourceTable.DefaultView.ToTable(false, cols);
//output.ProjectionSourceDataSource = projSource;
DataTable independentProj = projSource.DefaultView.ToTable(false, lra.Inputs);
DataTable dependentProj = projSource.DefaultView.ToTable(false, lra.Output);
double[][] input = independentProj.ToJagged();
double[] output1;
output1 = mlr.Regression.Transform(input);
DataTable result = projSource.Clone();
for (int i = 0; i < input.Length; i++)
{
DataRow row = result.NewRow();
for (int j = 0; j < lra.Inputs.Length; j++)
{
row[lra.Inputs[j]] = input[i][j];
}
row[lra.Output] = output1[i];
result.Rows.Add(row);
}
output.ProjectionResultDataSource = result;
var jsonResult = JsonConvert.SerializeObject(output, Formatting.Indented);
return(jsonResult);
}
catch (Exception ex)
{
throw new Exception("Error:" + ex);
}
}
