Package jline.io

Class Ret.pfqnMcmc

java.lang.Object

jline.io.Ret.pfqnMcmc

Enclosing class:

Ret

public static class Ret.pfqnMcmc
extends Object

Return of Pfqn_mcmc: the Chen-O'Cinneide regularization estimates and their batch-means confidence intervals.

There is deliberately NO normalizing constant here. The estimator is a ratio of holding-time weighted averages that yields G(N-e_r)/G(N) directly; G itself never enters the algorithm.

Field Summary

Fields

Modifier and Type	Field	Description
int	batches	batches the run was split into.
long	burnin	completions discarded as warm-up.
Matrix	Q	(M x R) mean queue lengths at the queueing stations.
Matrix	Qhi	(M x R) upper end of the two-sigma interval for Q.
Matrix	Qlo	(M x R) lower end of the two-sigma interval for Q.
Matrix	Qse	(M x R) batch-means standard error of Q.
long	samples	service completions actually simulated after warm-up.
Matrix	X	(1 x R) throughput estimates G(N-e_r)/G(N).
Matrix	Xhi	(1 x R) upper end of the two-sigma interval for X.
Matrix	Xlo	(1 x R) lower end of the two-sigma interval for X.
Matrix	Xse	(1 x R) batch-means standard error of X.

Constructor Summary

Constructors

Constructor	Description
pfqnMcmc(Matrix X, Matrix Q, Matrix Xse, Matrix Xlo, Matrix Xhi, Matrix Qse, Matrix Qlo, Matrix Qhi, int batches, long samples, long burnin)

Method Summary

Methods inherited from class java.lang.Object

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

Field Details

X

public Matrix X

(1 x R) throughput estimates G(N-e_r)/G(N).

Q

public Matrix Q

(M x R) mean queue lengths at the queueing stations.

Xse

public Matrix Xse

(1 x R) batch-means standard error of X.

Xlo

public Matrix Xlo

(1 x R) lower end of the two-sigma interval for X.

Xhi

public Matrix Xhi

(1 x R) upper end of the two-sigma interval for X.

Qse

public Matrix Qse

(M x R) batch-means standard error of Q.

Qlo

public Matrix Qlo

(M x R) lower end of the two-sigma interval for Q.

Qhi

public Matrix Qhi

(M x R) upper end of the two-sigma interval for Q.

batches

public int batches

batches the run was split into.

samples

public long samples

service completions actually simulated after warm-up.

burnin

public long burnin

completions discarded as warm-up.

Constructor Details

pfqnMcmc

public pfqnMcmc(Matrix X,
 Matrix Q,
 Matrix Xse,
 Matrix Xlo,
 Matrix Xhi,
 Matrix Qse,
 Matrix Qlo,
 Matrix Qhi,
 int batches,
 long samples,
 long burnin)