Package jline.solvers

Class NetworkMomentStationResult

java.lang.Object

jline.solvers.NetworkMomentStationResult

public class NetworkMomentStationResult
extends Object

Raw moment results behind NetworkMomentStationTable, mirroring the mom struct that MATLAB's getMomentStationTable returns as its second output.

Carries the per-station-total moments together with the cross-station covariance matrix Cov that the table does not show, and the underlying Pfqn_sens_mom (exact) or Pfqn_sens_linearizer (approximate) result.

See Also:

• NetworkMomentStationTable
• NetworkSolver.getMomentStationTable(int[])

Field Summary

Fields

Modifier and Type	Field	Description
Matrix	Cov	M x M, Cov(i,j) = Cov[Q_i,Q_j]; the diagonal is Var.
double	CovAsym	Asymmetry residual of the underlying recursion.
Matrix[][]	CovG	The general (M x G x M x G) covariance, CovG[i][g] being an M x G matrix whose (j,g2) entry is Cov[Q_(i,g),Q_(j,g2)].
Ret.pfqnSensMom	exact	The exact result, non-null when the method was "exact".
Ret.pfqnSensLinearizer	linearizer	The Linearizer result, non-null when the method was "lin".
Matrix	m	M x 1, m(i) = E[Q_i], the mean total queue length at station i.
Matrix	M2	M x 1, E[Q_i^2].
Matrix	M3	M x 1, E[Q_i^3].
Matrix	Skew	M x 1, skewness of Q_i; NaN where the variance vanishes.
Matrix	Var	M x 1, Var[Q_i].

Constructor Summary

Constructors

Constructor	Description
NetworkMomentStationResult()

Method Summary

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

exact

public Ret.pfqnSensMom exact

The exact result, non-null when the method was "exact".

linearizer

public Ret.pfqnSensLinearizer linearizer

The Linearizer result, non-null when the method was "lin".

m

public Matrix m

M x 1, m(i) = E[Q_i], the mean total queue length at station i.

Var

public Matrix Var

M x 1, Var[Q_i].

Cov

public Matrix Cov

M x M, Cov(i,j) = Cov[Q_i,Q_j]; the diagonal is Var. Non-null only in the single-group case, i.e. always for the station table and for the chain table of a single-chain model, mirroring the collapse of the MATLAB reference.

CovG

public Matrix[][] CovG

The general (M x G x M x G) covariance, CovG[i][g] being an M x G matrix whose (j,g2) entry is Cov[Q_(i,g),Q_(j,g2)]. Populated whenever the exact path ran, and the only form available when there is more than one group (more than one chain), where cross-chain and cross-station covariances both exist. Null on the Linearizer path, which approximates station totals only.

M2

public Matrix M2

M x 1, E[Q_i^2].

M3

public Matrix M3

M x 1, E[Q_i^3].

Skew

public Matrix Skew

M x 1, skewness of Q_i; NaN where the variance vanishes.

CovAsym

public double CovAsym

Asymmetry residual of the underlying recursion. Roundoff for the exact method; a genuine measure of approximation error for the Linearizer, whose fixed point does not enforce the symmetry the product form guarantees.

Constructor Details

NetworkMomentStationResult

public NetworkMomentStationResult()