WeightedLS Class |
Weighted Least Squares (WLS).
http://en.wikipedia.org/wiki/Weighted_least_squares#Weighted_least_squares
Inheritance HierarchyNamespace: FinMath.LeastSquares
Assembly: FinMath (in FinMath.dll) Version: 2.4.7-0a995bd0ea1854c2c868ec3f8dae606c5777e170
SyntaxC#
[SerializableAttribute] public class WeightedLS : AbstractLS
The WeightedLS type exposes the following members.
Constructors| Name | Description | |
|---|---|---|
 | WeightedLS(Int32) |
Initialize new instance of WLS.
|
| WeightedLS(Matrix, Vector, Vector) |
Initialize new instance of WLS.
Note: That weights commonly used as INVERTED residuals variance, NOT VARIANCE.
Use WeightedLS.VariancesToWeights to robust conversion variance to weight.
|
Properties| Name | Description | |
|---|---|---|
 | Parameters |
Vector of regression parameters.
(Inherited from AbstractLS.) |
| Status |
Method Status.
(Inherited from AbstractLS.) |
Methods| Name | Description | |
|---|---|---|
| CalculateR2(Matrix, Vector) |
Calculate R2 - Coefficient of determination - goodness-of-fit of the OLS regression metric.
Method use computed regression parameters.
http://en.wikipedia.org/wiki/Coefficient_of_determination
(Inherited from AbstractLS.) |
| Equals | Determines whether the specified object is equal to the current object. (Inherited from Object.) |
| Estimate(Matrix) |
Estimate regresand values, using computed regression parameters and regressors values.
Use this this method for multiple observations.
(Inherited from AbstractLS.) |
| Estimate(Vector) |
Estimate regresand value, using computed regression parameters and regressors value.
Use this this method for single observation.
(Inherited from AbstractLS.) |
| EstimateResidual(Vector, Double) |
Calculate residuals of estimations, using computed regression parameters, regressors and regressand value.
Use this this method for single observation.
(Inherited from AbstractLS.) |
| EstimateResiduals(Matrix, Vector) |
Calculate residuals of estimations, using computed regression parameters, regressors and regressand values.
Use this this method for multiple observations.
(Inherited from AbstractLS.) |
 | FitWLS |
Static method, which can be used to simply calculate WLS parameters.
Note: That weights commonly used as INVERTED residuals variance, NOT VARIANCE.
Use WeightedLS.VariancesToWeights to robust conversion variance to weight.
|
| Forget(Matrix, Vector, Vector) |
Update WLS parameters according new observations.
It is useful for estimation on sliding window.
|
| Forget(Vector, Double, Double) |
Update WLS parameters according new observations.
It is useful for estimation on sliding window.
|
| GetHashCode | Serves as the default hash function. (Inherited from Object.) |
| GetType | Gets the Type of the current instance. (Inherited from Object.) |
| ToString | Returns a string that represents the current object. (Inherited from Object.) |
| Update(Matrix, Vector, Vector) |
Update WLS parameters according new observations.
It is useful for estimation on sliding window.
|
| Update(Vector, Double, Double) |
Update WLS parameters according new observations.
It is useful for estimation on sliding window.
|
| VariancesToWeights(Vector) |
Static method which transform known residuals variance to weights,
Note: We for zero variances we assume average weights.
|
| VariancesToWeights(Vector, Vector) |
Static method which transform known residuals variance to weights,
Note: We for zero variances we assume average weights.
|
| VariancesToWeights(Vector, Vector, Double) |
Static method which transform known residuals variance to weights,
Note: We for zero variances we assume average weights.
|
See Also