×

Fitting matrix geometric distributions by model reduction. (English) Zbl 1328.62227

Summary: A novel algorithmic method is proposed to fit matrix geometric distributions of desired order to empirical data or arbitrary discrete distributions. The proposed method effectively combines two existing approaches from two different disciplines: well-established model reduction methods used in system theory and moment matching methods of applied probability that employ second-order discrete phase-type distributions. The proposed approach is validated with exhaustive numerical examples including well-known statistical data.

MSC:

62G07 Density estimation
78M34 Model reduction in optics and electromagnetic theory
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] Neuts M.F., Matrix-Geometric Solutions in Stochastic Models: An Algorithmic Approach (1981) · Zbl 0469.60002
[2] DOI: 10.1137/1.9780898719734 · Zbl 0922.60001 · doi:10.1137/1.9780898719734
[3] Nielsen, B.F. Modelling of multiple access systems with phase type distributions. Ph.D. dissertation, IMSOR, Technical University of Denmark, 1988.
[4] DOI: 10.1007/BF01721011 · Zbl 0463.60073 · doi:10.1007/BF01721011
[5] DOI: 10.1080/03461230110106471 · Zbl 1142.62088 · doi:10.1080/03461230110106471
[6] Neuts M.F., Liber Amicorum Professor Emeritus H. Florin pp 173– (1975)
[7] DOI: 10.1214/aop/1176993932 · Zbl 0479.60085 · doi:10.1214/aop/1176993932
[8] DOI: 10.1080/15326349908807134 · Zbl 0701.62021 · doi:10.1080/15326349908807134
[9] DOI: 10.1214/aop/1176989805 · Zbl 0755.60049 · doi:10.1214/aop/1176989805
[10] Asmussen S., Encyclopedia of Statistical Sciences pp 435– (1997)
[11] Fackrell, M.W. Characterization of matrix-exponential distributions. Ph.D. dissertation, University of Adelaide, 1993. · Zbl 1197.60013
[12] DOI: 10.1239/aap/1175266478 · Zbl 1114.60013 · doi:10.1239/aap/1175266478
[13] Asmussen S., Matrix Analytic Methods in Stochastic Models pp 313– (1997)
[14] DOI: 10.1016/j.peva.2004.12.002 · doi:10.1016/j.peva.2004.12.002
[15] DOI: 10.1016/j.peva.2010.08.023 · Zbl 05842032 · doi:10.1016/j.peva.2010.08.023
[16] DOI: 10.1007/978-1-4419-9070-9 · doi:10.1007/978-1-4419-9070-9
[17] Greeuw, S. On the relation between matrix-geometric and discrete phase-type distributions. Master’s thesis, University of Amsterdam, 2009.
[18] DOI: 10.1080/15326349108807207 · Zbl 0741.60011 · doi:10.1080/15326349108807207
[19] DOI: 10.1145/1190366.1190377 · doi:10.1145/1190366.1190377
[20] DOI: 10.1016/j.peva.2013.04.003 · doi:10.1016/j.peva.2013.04.003
[21] Asmussen S., Symposium i Anvendt Statistik pp 333– (1991)
[22] DOI: 10.1007/3-540-46029-2_5 · doi:10.1007/3-540-46029-2_5
[23] DOI: 10.1109/TDSC.2006.27 · Zbl 05341111 · doi:10.1109/TDSC.2006.27
[24] DOI: 10.1016/j.camwa.2012.03.016 · Zbl 1268.62059 · doi:10.1016/j.camwa.2012.03.016
[25] DOI: 10.1007/s10092-007-0139-7 · Zbl 1168.62313 · doi:10.1007/s10092-007-0139-7
[26] DOI: 10.1080/15326349908807147 · Zbl 0708.60019 · doi:10.1080/15326349908807147
[27] Telek M., Int. J. Simulation 3 pp 47– (2002)
[28] Papp J., Proceedings of UK Performance Evaluation Workshop (UKPEW 2013) (2013)
[29] DOI: 10.1080/15326340701300712 · Zbl 1130.60077 · doi:10.1080/15326340701300712
[30] DOI: 10.1081/STM-200056227 · Zbl 1065.62097 · doi:10.1081/STM-200056227
[31] Kailath T., Linear Systems (1980)
[32] DOI: 10.1109/TAC.1981.1102568 · Zbl 0464.93022 · doi:10.1109/TAC.1981.1102568
[33] DOI: 10.1090/conm/280/04630 · doi:10.1090/conm/280/04630
[34] DOI: 10.1016/j.laa.2004.12.022 · Zbl 1112.93015 · doi:10.1016/j.laa.2004.12.022
[35] DOI: 10.1016/S0168-9274(02)00116-2 · Zbl 1012.65136 · doi:10.1016/S0168-9274(02)00116-2
[36] DOI: 10.1016/j.laa.2007.10.041 · Zbl 1137.93006 · doi:10.1016/j.laa.2007.10.041
[37] Grussler, C. Model reduction of positive systems. Ph.D. dissertation, Technical University of Kaiserslautern, 2012.
[38] DOI: 10.1016/j.automatica.2011.02.032 · Zbl 1220.93036 · doi:10.1016/j.automatica.2011.02.032
[39] Cha S.-H., Int. J. Math. Models Methods Appl. Sci. 1 pp 300– (2007)
[40] DOI: 10.1080/10485250701733747 · Zbl 1130.62030 · doi:10.1080/10485250701733747
[41] DOI: 10.1023/A:1025653200299 · Zbl 1046.90016 · doi:10.1023/A:1025653200299
[42] Chen C.-T., Linear System Theory and Design (1998)
[43] DOI: 10.1080/10637190208941436 · Zbl 1009.65039 · doi:10.1080/10637190208941436
[44] Golub G.H., Matrix Computations (1996) · Zbl 0865.65009
[45] DOI: 10.1137/S1064827598347666 · Zbl 0958.65052 · doi:10.1137/S1064827598347666
[46] DOI: 10.1016/j.laa.2006.01.007 · Zbl 1092.65053 · doi:10.1016/j.laa.2006.01.007
[47] DOI: 10.1023/A:1022205420182 · Zbl 1034.93020 · doi:10.1023/A:1022205420182
[48] DOI: 10.2307/2347385 · Zbl 0707.62186 · doi:10.2307/2347385
[49] Cho K., USENIX 2000 Annual Technical Conference: FREENIX Track pp 263– (2000)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.