private static double GetLikelihood_Normal(List <RDataRecord> data, double[] beta_hat)
{
int N = data.Count;
int k = beta_hat.Length;
double residual_sum_of_squares = 0;
double[] y = new double[N];
for (int i = 0; i < N; ++i)
{
y[i] = data[i].YValue;
}
double sigma = StdDev.GetStdDev(y, Mean.GetMean(y)) / (N - k - 1);
for (int i = 0; i < N; ++i)
{
double linear_predictor = 0;
RDataRecord rec = data[i];
for (int j = 0; j < k; ++j)
{
linear_predictor += rec.data[j] * beta_hat[j];
}
double residual = rec.YValue - linear_predictor;
residual_sum_of_squares += residual * residual;
}
return(System.Math.Exp(-residual_sum_of_squares / (2 * sigma)) / System.Math.Sqrt(2 * System.Math.PI * sigma));
}
public virtual Glm CreateSolver(IList <RDataRecord> records)
{
RDataRecord rec = records[0];
int d = rec.Dimension;
int m = records.Count;
int n = 0;
Dictionary <int, Dictionary <int, int> > levelOrg;
Analyze(records, out levelOrg, out n);
double[,] A = new double[m, n];
double[] b = new double[m];
for (int i = 0; i < m; ++i)
{
rec = records[i];
int index = 0;
for (int j = 0; j < d; ++j)
{
A[i, index++] = rec.data[j];
}
b[i] = rec.YValue;
}
return(new GlmIrlsQrNewton(mDistributionFamily, A, b));
}
private static void Analyze(IList <RDataRecord> records, out Dictionary <int, Dictionary <int, int> > levelOrg, out int n)
{
RDataRecord rec = records[0];
int m = records.Count;
n = 1;
levelOrg = new Dictionary <int, Dictionary <int, int> >();
for (int i = 1; i < n; ++i)
{
n++;
}
}
public static double Predict(Glm solver, IList <RDataRecord> records, RDataRecord input_0)
{
int d = input_0.Dimension;
int n = 0;
Dictionary <int, Dictionary <int, int> > levelOrg;
Analyze(records, out levelOrg, out n);
double[] x = new double[n];
int index = 0;
for (int j = 0; j < d; ++j)
{
x[index++] = input_0.data[j];
}
return(solver.Predict(x));
}