| FloatLeastSquares(FloatMatrix, FloatVector, Boolean, Single) Constructor |
Constructs a least squares solution for the given linear system
Ax = y, optionally adding an intercept parameter, and using
the specified tolerance to compute the effective rank.
Namespace: CenterSpace.NMath.CoreAssembly: NMath (in NMath.dll) Version: 7.4
Syntax public FloatLeastSquares(
FloatMatrix A,
FloatVector y,
bool addIntercept,
float tolerance
)
Public Sub New (
A As FloatMatrix,
y As FloatVector,
addIntercept As Boolean,
tolerance As Single
)
public:
FloatLeastSquares(
FloatMatrix^ A,
FloatVector^ y,
bool addIntercept,
float tolerance
)
new :
A : FloatMatrix *
y : FloatVector *
addIntercept : bool *
tolerance : float32 -> FloatLeastSquares
Parameters
- A FloatMatrix
- The matrix.
- y FloatVector
- The right hand side.
- addIntercept Boolean
- If true, a column of ones is prepended to
a deep copy of matrix A before solving for the least squares solution.
This corresponds to adding an intercept parameter to the model. Note: the
input matrix A is not changed.
- tolerance Single
- Tolerance used to compute the effective rank
of A The effective rank of A is defined as the order of the
largest leading submatrix, R11, in the QR factorization with pivoting of A,
whose estimated condition number is less than 1/tolerance.
Exceptions Remarks
The effective rank of A is determined by treating as zero
those singular values that are less than tolerance times the largest
singular value.
See Also