Filters








23 Hits in 3.3 sec

On Bayesian inference for the M/G/1 queue with efficient MCMC sampling [article]

Alexander Y. Shestopaloff, Radford M. Neal
2014 arXiv   pre-print
We introduce an efficient MCMC sampling scheme to perform Bayesian inference in the M/G/1 queueing model given only observations of interdeparture times.  ...  We show that our novel updates improve the speed of sampling considerably, by factors of about 60 to about 180 on a variety of simulated data sets.  ...  Numerical values of the simulated interdeparture times are presented in Table 1 .  ... 
arXiv:1401.5548v1 fatcat:gk4bei65azglfmfenquppl74um

Characterization of the departure process from anME/ME/1queue

Jayesh Kumaran, Kenneth Mitchell, Appie van de Liefvoort
2004 Reserche operationelle  
The approximations capture the interdeparture times from an ME/ME/1 queue exactly and preserve the lag correlations of inter-event times of the departures from an ME/ME/1 queue up to lag (k − 1).  ...  In this paper we propose a family of finite approximations for the departure process of an ME/ME/1 queue indexed by a parameter k defined as the system size of the finite approximation.  ...  transformation of the arrival time distribution.  ... 
doi:10.1051/ro:2004018 fatcat:lyecma3xfndqlcrtemzq5j2yki

Correlated phase-type distributed random numbers as input models for simulations

Jan Kriege, Peter Buchholz
2011 Performance evaluation (Print)  
Examples for correlated events are the traffic in computer networks [30] , access times and locations at a disk in a computer [32] and also failure times of software [20] .  ...  In this paper, ARTA processes which have been used several times for describing correlated input processes in simulation are extended by using ARMA instead of AR processes to realize the correlation and  ...  case interdeparture times of packets in a computer network, then the likelihood function is given by L(T, X) = l i=1 f X (t i ).  ... 
doi:10.1016/j.peva.2011.07.014 fatcat:3ts2c4ku2zbpldamp76lmc5wpi

Semi Markov Process [chapter]

Lester Lipsky
2008 Queueing Theory  
In this chapter we set up a general procedure for creating a sequence of random variables {X i } which may be thought of as the interarrival times of successive customers for some arrival process.  ...  The set of formulas for the joint interdeparture distributions and correlation lag-k number are then set up, thereby showing that, indeed, the X i and X j are in most cases, correlated.  ...  Second, the inverses of matrices are found by inverting matrices of smaller dimension. Third, one can go further by analytic manipulation rather than having to resort to numerical computation.  ... 
doi:10.1007/978-0-387-49706-8_8 fatcat:jcd75ohh7jblldudhttc7quqxq

Modeling Stochastic Lead Times in Multi-Echelon Systems

E. B. Diks, M. C. van der Heijden
1997 Probability in the engineering and informational sciences (Print)  
In this paper, we use results from queuing theory to define a set of simple lead time processes guaranteeing that (a) orders do not cross and (b) prespecified means and variances of all lead times in the  ...  In many multi-echelon inventory systems, the lead times are random variables.  ...  The variance of the interdeparture time of stockpoint 1 can be determined from the three approximations for the squared coefficient of variation of the interdeparture times of a GI/G/1 queue given by Buzacott  ... 
doi:10.1017/s0269964800004988 fatcat:gifoybnz6raxvmzn3gbu7fsxwm

The Consequences of Time-Phased Order Releases on Two M/M/1 Queues in Series

Diederik Claerhout, Nico Vandaele
2009 Social Science Research Network  
It is a process characterized by a negligible autocorrelation.  ...  A key characteristic of MRP applications includes the coordination of assembly and purchased component requirements by time-phased order releases.  ...  In the next section, we study the autocorrelation in the interdeparture times at the "Wait for release" bu¤er.  ... 
doi:10.2139/ssrn.1512483 fatcat:t27predtwrhrjfwhosksmgrx6a

Data Provenance and Management in Radio Astronomy: A Stream Computing Approach [chapter]

Mahmoud S. Mahmoud, Andrew Ensor, Alain Biem, Bruce Elmegreen, Sergei Gulyaev
2013 Studies in Computational Intelligence  
We present a case-study: the InfoSphere Streams implementation of an autocorrelating spectrometer, and us-  ...  It is a shift from conventional data mining techniques (involving analysis of existing data from databases) towards real-time analytic processing.  ...  Acknowledgements The first author would like to thank the New Zealand government's Tertiary Education Commission Build IT fund for funding this research and the IBM cooperation for providing the computing  ... 
doi:10.1007/978-3-642-29931-5_6 fatcat:j37dadnqibd7leyxi25fyvl25m

Data Provenance and Management in Radio Astronomy: A Stream Computing Approach [article]

Mahmoud S. Mahmoud, Andrew Ensor, Alain Biem, Bruce Elmegreen and Sergei Gulyaev
2011 arXiv   pre-print
We present a case-study: the InfoSphere Streams implementation of an autocorrelating spectrometer, and using this example we discuss the advantages of the stream-computing approach and the utilization  ...  It is a shift from conventional data mining techniques (involving analysis of existing data from databases) towards real-time analytic processing.  ...  Acknowledgements The first author would like to thank the New Zealand government's Tertiary Education Commission Build IT fund for funding this research and the IBM cooperation for providing the computing  ... 
arXiv:1112.2584v1 fatcat:uxycuqxl3zbwpgsfxltlvico4q

A unified ME algorithm for arbitrary open QNMs with mixed blocking mechanisms

Kevin Smith, Jumma Alanazi, Demetres Kouvatsos
2011 Numerical Algebra, Control and Optimization  
determine the first two moments of BAS-and BBS-dependent effective service times towards a node-by-node decomposition of the entire network.  ...  This system of queues is employed to determine new BAS-and BBS-based analytic expressions for the first two moments of the BM-dependent effective service times and overall interarrival times.  ...  The main computational cost of the ME algorithm is proportional to the inversion of the associated Jacobian matrix of the non-linear system of equations for the blocking probabilities { }.  ... 
doi:10.3934/naco.2011.1.781 fatcat:aavlzxlr2bhlvptjvoq4l74avy

A Unified ME Algorithm for Arbitrary Open QNMs with Mixed Blocking Mechanisms

Jumma S. Alanazi, Demetres D. Kouvatsos
2011 2011 IEEE/IPSJ International Symposium on Applications and the Internet  
determine the first two moments of BAS-and BBS-dependent effective service times towards a node-by-node decomposition of the entire network.  ...  This system of queues is employed to determine new BAS-and BBS-based analytic expressions for the first two moments of the BM-dependent effective service times and overall interarrival times.  ...  The main computational cost of the ME algorithm is proportional to the inversion of the associated Jacobian matrix of the non-linear system of equations for the blocking probabilities { }.  ... 
doi:10.1109/saint.2011.91 dblp:conf/saint/AlanaziK11 fatcat:7dxhjn2xtba23nxmpaxrssg2tq

Bandwidth allocation under end-to-end percentile delay bounds

Bushra Anjum, Harry Perros, Xenia Mountrouidou, Kimon Kontovasilis
2011 International Journal of Network Management  
The second algorithm is based on traces, such as video traces, that consist of a sequence of tuplets, each describing the time of arrival of a packet and its packet length.  ...  The path through an IP network is depicted by a tandem queueing network of infinite capacity queues, where each queue represents the output port of a router.  ...  The last method discussed involves the numerical inversion of the Laplace transforms following Abate and Whitt [28] .  ... 
doi:10.1002/nem.783 fatcat:z3s5maemznfqfnlklu5rv7rssq

Page 1952 of Mathematical Reviews Vol. , Issue Index [page]

Mathematical Reviews  
Computation of autocorrelations of interdeparture times by numerical transform inversion. (English summary ) — On the numerical inversion of busy-period related transforms.  ...  Bayesian regularization of numerical inversion of Mellin transform.  ... 

Neural Density Estimation and Likelihood-free Inference [article]

George Papamakarios
2019 arXiv   pre-print
I consider two problems in machine learning and statistics: the problem of estimating the joint probability density of a collection of random variables, known as density estimation, and the problem of  ...  The contribution of the thesis is a set of new methods for addressing these problems that are based on recent advances in neural networks and deep learning.  ...  George Papamakarios and Theo Pavlakou were supported by the Centre for Doctoral Training in Data Science, funded by EPSRC (grant EP/L016427/1) and the University of Edinburgh.  ... 
arXiv:1910.13233v1 fatcat:ftbqx3mno5e4bdxiufizdcglhq

Constructing Summary Statistics for Approximate Bayesian Computation: Semi-automatic ABC [article]

Paul Fearnhead, Dennis Prangle
2011 arXiv   pre-print
Approximate Bayesian computation (ABC) is a method of inference for such models.  ...  It replaces calculation of the likelihood by a step which involves simulating artificial data for different parameter values, and comparing summary statistics of the simulated data to summary statistics  ...  method of Ripley (1987) (page 98) and performing inversion by substituting these in (10).  ... 
arXiv:1004.1112v2 fatcat:y4epxkwrazcktl5fckye7o7seq

Approximate Bayesian computation with the Wasserstein distance

Espen Bernton, Pierre E. Jacob, Mathieu Gerber, Christian P. Robert
2019 Journal of The Royal Statistical Society Series B-statistical Methodology  
A growing number of generative statistical models do not permit the numerical evaluation of their likelihood functions.  ...  Various extensions to time series data are discussed.  ...  Jacob gratefully acknowledges support by the National Science Foundation through grant DMS-1712872.  ... 
doi:10.1111/rssb.12312 fatcat:ksct7kdr6jhslklldwvd2g3gjm
« Previous Showing results 1 — 15 out of 23 results