×

Steady state probability vector of positive definite regularized linear systems of circulant stochastic matrices. (English) Zbl 1360.65099

Let \(Q\) be a square circulant matrix whose row and column sums are zero. The problem of solving the regularized linear system \(Ax=b\) is considered, where \(A=Q^T+\epsilon I\), with \(\epsilon > 0\). Let \(D\) be the diagonal matrix whose diagonal entries are the corresponding diagonal entries of \(A\) and whose lower and upper parts give rise to the lower triangular matrix \(L\) and the upper triangular matrix \(U\), respectively. Here, the diagonal entries of \(L\) and \(U\) are all equal \(\epsilon /2\). Then \(A=T+S\), where \(T\) is a triangular matrix with nonnegative diagonal elements and \(S\) is a symmetric matrix with positive diagonal entries and negative off-diagonal elements. Such a splitting is referred to as a \(TS\)-splitting. It is shown that \(A\) is positive definite. An iterative scheme is proposed using the \(TS\)-splitting. It is shown that this scheme converges to the unique solution of the regularized system. Comparisons are made with the existing methods for a numerical example.

MSC:

65F10 Iterative numerical methods for linear systems
15B51 Stochastic matrices
65F22 Ill-posedness and regularization problems in numerical linear algebra
Full Text: DOI

References:

[1] DOI: 10.1109/90.282603 · doi:10.1109/90.282603
[2] DOI: 10.1109/90.392383 · doi:10.1109/90.392383
[3] DOI: 10.1109/90.650143 · doi:10.1109/90.650143
[4] DOI: 10.1109/5.650182 · doi:10.1109/5.650182
[5] DOI: 10.1137/0607044 · Zbl 0617.65027 · doi:10.1137/0607044
[6] DOI: 10.1023/A:1016616406118 · Zbl 1030.68906 · doi:10.1023/A:1016616406118
[7] Kasahara S, IEICE Trans. Commun 84 pp 2134– (2001)
[8] DOI: 10.1093/ietcom/e88-b.4.1493 · doi:10.1093/ietcom/e88-b.4.1493
[9] DOI: 10.1109/JSAC.1986.1146393 · doi:10.1109/JSAC.1986.1146393
[10] DOI: 10.1016/0166-5316(93)90035-S · Zbl 0781.60098 · doi:10.1016/0166-5316(93)90035-S
[11] DOI: 10.1080/15326348908807110 · Zbl 0673.60102 · doi:10.1080/15326348908807110
[12] DOI: 10.4236/am.2014.53050 · doi:10.4236/am.2014.53050
[13] Neuts MF, Matrix-geometric solutions in stochastic models: an algorithmic approach (1981)
[14] Stewart WJ, Introduction to numerical solution of Markov chains (1994)
[15] DOI: 10.1007/BF01189232 · Zbl 0789.60067 · doi:10.1007/BF01189232
[16] DOI: 10.1002/(SICI)1099-1506(199807/08)5:4<299::AID-NLA103>3.0.CO;2-V · Zbl 0937.65038 · doi:10.1002/(SICI)1099-1506(199807/08)5:4<299::AID-NLA103>3.0.CO;2-V
[17] DOI: 10.1287/opre.40.6.1156 · Zbl 0764.65095 · doi:10.1287/opre.40.6.1156
[18] DOI: 10.1137/1.9780898718003 · doi:10.1137/1.9780898718003
[19] DOI: 10.1002/nla.821 · Zbl 1265.60146 · doi:10.1002/nla.821
[20] DOI: 10.1137/070687888 · Zbl 1194.65052 · doi:10.1137/070687888
[21] DOI: 10.1137/040616541 · Zbl 1097.65047 · doi:10.1137/040616541
[22] DOI: 10.1137/S0895479894262339 · Zbl 0876.65019 · doi:10.1137/S0895479894262339
[23] DOI: 10.1002/nla.1680010406 · Zbl 0840.65021 · doi:10.1002/nla.1680010406
[24] Hayami K, Proceedings of Fifth China–Japan Seminar on Numerical Mathematics, Shanghai, 2000 pp 117– (2002)
[25] DOI: 10.1007/BF02481090 · Zbl 0569.60069 · doi:10.1007/BF02481090
[26] Benzi M, Electron. Trans. Numer. Anal 26 pp 209– (2007)
[27] Benzi M, Proceedings of the 2006 Markov Anniversary Meeting pp 239– (2006)
[28] DOI: 10.1007/s10444-011-9262-8 · Zbl 1323.65030 · doi:10.1007/s10444-011-9262-8
[29] DOI: 10.1016/j.camwa.2011.09.041 · Zbl 1236.65009 · doi:10.1016/j.camwa.2011.09.041
[30] DOI: 10.1137/S1064827503428114 · Zbl 1079.65028 · doi:10.1137/S1064827503428114
[31] DOI: 10.1137/S0895479801395458 · Zbl 1036.65032 · doi:10.1137/S0895479801395458
[32] DOI: 10.1007/s002110000173 · Zbl 0964.65034 · doi:10.1007/s002110000173
[33] DOI: 10.1016/j.cam.2008.10.033 · Zbl 1184.65038 · doi:10.1016/j.cam.2008.10.033
[34] Minc H, Nonnegative matrices (1988)
[35] DOI: 10.1137/1.9781611971262 · doi:10.1137/1.9781611971262
[36] DOI: 10.1137/0910004 · Zbl 0666.65029 · doi:10.1137/0910004
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.