Package jline.api.pfqn.mva
Class Pfqn_mvajd
java.lang.Object
jline.api.pfqn.mva.Pfqn_mvajd
The two names denote the SAME routine because the recursion evaluates the rate
handle at a full occupancy vector, mu_i(s_i + e_r) with s_i the shift (the
occupancy already committed at the bottom of station i), and never inspects the
structure of mu_i. That is exactly the "third form" of the Conditional MVA of
Casale (QUESTA 2009), a rate depending on the full per-class occupancy vector,
so any joint-dependent scaling eta_i(n) (sn.jdscaling) is admissible.
Unlike the AMVA joint-dependence route (Solver_amvald with sn.jdscaling, which evaluates eta at the MEAN arrival-instant vector 1 + E[Q] and therefore collapses a support indicator to 1), this routine evaluates the rate at exact integer occupancies and is exact for the balanced-fair station, at the cost of walking prod_r C(N_r+K+1,K+1) states with K joint-dependent stations.
-
Method Summary
Modifier and TypeMethodDescriptionstatic Pfqn_mvaoi.Resultpfqn_mvajd(double[] Z, int[] N, ToDoubleFunction<int[]> mu) static Pfqn_mvaoi.Resultpfqn_mvajd(double[] Z, int[] N, List<ToDoubleFunction<int[]>> mu, double[][] Dli) static Pfqn_mvaoi.Resultpfqn_mvajd(double[] Z, int[] N, List<ToDoubleFunction<int[]>> mu, double[][] Dli, double[][] visits)
-
Method Details
-
pfqn_mvajd
-
pfqn_mvajd
public static Pfqn_mvaoi.Result pfqn_mvajd(double[] Z, int[] N, List<ToDoubleFunction<int[]>> mu, double[][] Dli) -
pfqn_mvajd
public static Pfqn_mvaoi.Result pfqn_mvajd(double[] Z, int[] N, List<ToDoubleFunction<int[]>> mu, double[][] Dli, double[][] visits)
-