| MultipleRegressionDirectMethodT Method (IEnumerableTupleT, T, Boolean, DirectRegressionMethod) |
Find the model parameters β such that their linear combination with all predictor-arrays in X become as close to their response in Y as possible, with least squares residuals.
Uses the cholesky decomposition of the normal equations.
Namespace:
MathNet.Numerics.LinearRegression
Assembly:
MathNet.Numerics (in MathNet.Numerics.dll) Version: 3.7
Syntax public static T[] DirectMethod<T>(
IEnumerable<Tuple<T[], T>> samples,
bool intercept = false,
DirectRegressionMethod method = DirectRegressionMethod.NormalEquations
)
where T : struct, new(), IEquatable<T>, IFormattable
Parameters
- samples
- Type: System.Collections.GenericIEnumerableTupleT, T
Sequence of predictor-arrays and their response. - intercept (Optional)
- Type: SystemBoolean
True if an intercept should be added as first artificial predictor value. Default = false. - method (Optional)
- Type: MathNet.Numerics.LinearRegressionDirectRegressionMethod
The direct method to be used to compute the regression.
Type Parameters
- T
Return Value
Type:
TBest fitting list of model parameters β for each element in the predictor-arrays.
See Also