Gets the regression coefficients matrix, B, for the PLS2 calculation.
B satisifies the relationship
CopyC#
where X and ResponseVector are, respectively, the centered independent
and dependent variables values, and E is a noise term for the model.
Namespace:
CenterSpace.NMath.StatsResponseVector = XB + E
Assembly: NMathStats (in NMathStats.dll) Version: 3.4.0.0
Syntax
| C# |
|---|
public DoubleMatrix Coefficients { get; } |
| Visual Basic (Declaration) |
|---|
Public ReadOnly Property Coefficients As DoubleMatrix |
| Visual C++ |
|---|
public: property DoubleMatrix^ Coefficients { DoubleMatrix^ get (); } |
Remarks
The coeffient matrix, B, can be used for prediction as follows:
Let yhat(z) be the predicted value for the independet variable
value z, xbar the mean of the PredictorMatrix variables and
ybar the mean of ResponseVector variables. Then
CopyC#yhat(z) = ybar + (z - xbar)B