Quasistationarity of continuous-time Markov chains with positive drift

We shall study continuous-time Markov chains on the nonnegative integers which are both irreducible and transient, and which exhibit discernible stationarity before drift to infinity sets in. We will show how this 'quasi' stationary behaviour can be modelled using a limiting conditional distribution: specifically, the limiting state probabilities conditional on not having left 0 for the last time. By way of a dual chain, obtained by killing the original process on last exit from 0, we invoke the theory of quasistationarity for absorbing Markov chains. We prove that the conditioned state probabilities of the original chain are equal to the state probabilities of its dual conditioned on non-absorption, thus allowing us to establish the simultaneous existence and then equivalence, of their limiting conditional distributions. Although a limiting conditional distribution for the dual chain is always a quasistationary distribution in the usual sense, a similar statement is not possible for the original chain.

Further results on the relationship between mu-invariant measures and quasi-stationary distributions for absorbing continuous-time Markov chains

This note considers continuous-time Markov chains whose state space consists of an irreducible class, C, and an absorbing state which is accessible from C. The purpose is to provide results on mu-invariant and mu-subinvariant measures where absorption occurs with probability less than one. In particular, the well-known premise that the mu-invariant measure, m, for the transition rates be finite is replaced by the more natural premise that m be finite with respect to the absorption probabilities. The relationship between mu-invariant measures and quasi-stationary distributions is discussed. (C) 2000 Elsevier Science Ltd. All rights reserved.

New methods for determining quasi-stationary distributions for Markov chains

We shall be concerned with the problem of determining quasi-stationary distributions for Markovian models directly from their transition rates Q. We shall present simple conditions for a mu-invariant measure m for Q to be mu-invariant for the transition function, so that if m is finite, it can be normalized to produce a quasi-stationary distribution. (C) 2000 Elsevier Science Ltd. All rights reserved.

Morphological and electrophysiological properties of principal neurons in the rat lateral amygdala in vitro

In this study, we characterize the electrophysiological and morphological properties of spiny principal neurons in the rat lateral amygdala using whole cell recordings in acute brain slices. These neurons exhibited a range of firing properties in response to prolonged current injection. Responses varied from cells that showed full spike frequency adaptation, spiking three to five times, to those that showed no adaptation. The differences in firing patterns were largely explained by the amplitude of the afterhyperpolarization (AHP) that followed spike trains. Cells that showed full spike frequency adaptation had large amplitude slow AHPs, whereas cells that discharged tonically had slow AHPs of much smaller amplitude. During spike trains, all cells showed a similar broadening of their action potentials. Biocytin-filled neurons showed a range of pyramidal-like morphologies, differed in dendritic complexity, had spiny dendrites, and differed in the degree to which they clearly exhibited apical versus basal dendrites. Quantitative analysis revealed no association between cell morphology and firing properties. We conclude that the discharge properties of neurons in the lateral nucleus, in response to somatic current injections, are determined by the differential distribution of ionic conductances rather than through mechanisms that rely on cell morphology.

Visualization of learning in multilayer perceptron networks using principal component analysis

This paper is concerned with the use of scientific visualization methods for the analysis of feedforward neural networks (NNs). Inevitably, the kinds of data associated with the design and implementation of neural networks are of very high dimensionality, presenting a major challenge for visualization. A method is described using the well-known statistical technique of principal component analysis (PCA). This is found to be an effective and useful method of visualizing the learning trajectories of many learning algorithms such as back-propagation and can also be used to provide insight into the learning process and the nature of the error surface.

Neighborhood and Efficiency in Manufacturing in Brazilian Regions: A Spatial Markov Chain Analysis

This paper analyzes the geography of regional competitiveness in manufacturing in Brazil. The authors estimate stochastic frontiers to calculate regional efficiency of representative firms in 137 regions in the period 2000-2006, in four sectors defined by technological intensity. The efficiency results are analyzed using Markov Spatial Transition Matrices to provide insights into the transition of regions between efficiency levels, considering their local spatial context. The results indicate that geography plays an important role in manufacturing competitiveness. In particular, regions with more competitive neighbors are more likely to improve their relative efficiency (pull effect) over time, and regions with less competitive neighbors are more likely to lose relative efficiency (drag effect). The authors find that the pull effect is stronger than the drag effect.

Characterization of the rat exploratory behavior in the elevated plus-maze with Markov chains

The elevated plus-maze is an animal model of anxiety used to study the effect of different drugs on the behavior of the animal It consists of a plus-shaped maze with two open and two closed arms elevated 50 cm from the floor The standard measures used to characterize exploratory behavior in the elevated plus-maze are the time spent and the number of entries in the open arms In this work we use Markov chains to characterize the exploratory behavior of the rat in the elevated plus-maze under three different conditions normal and under the effects of anxiogenic and anxiolytic drugs The spatial structure of the elevated plus-maze is divided into squares which are associated with states of a Markov chain By counting the frequencies of transitions between states during 5-min sessions in the elevated plus-maze we constructed stochastic matrices for the three conditions studied The stochastic matrices show specific patterns which correspond to the observed behaviors of the rat under the three different conditions For the control group the stochastic matrix shows a clear preference for places in the closed arms This preference is enhanced for the anxiogenic group For the anxiolytic group the stochastic matrix shows a pattern similar to a random walk Our results suggest that Markov chains can be used together with the standard measures to characterize the rat behavior in the elevated plus-maze (C) 2010 Elsevier B V All rights reserved

Strong Grobner bases for polynomials over a principal ideal ring

Grobner bases have been generalised to polynomials over a commutative ring A in several ways. Here we focus on strong Grobner bases, also known as D-bases. Several authors have shown that strong Grobner bases can be effectively constructed over a principal ideal domain. We show that this extends to any principal ideal ring. We characterise Grobner bases and strong Grobner bases when A is a principal ideal ring. We also give algorithms for computing Grobner bases and strong Grobner bases which generalise known algorithms to principal ideal rings. In particular, we give an algorithm for computing a strong Grobner basis over a finite-chain ring, for example a Galois ring.

CDKN2A is not the principal target of deletions on the short arm of chromosome 9 in neuroendocrine (Merkel cell) carcinoma of the skin

The majority of small-cell lung cancers (SCLCs) express p16 but not pRb, Given our previous study showing loss of pRb in Merkel cell carcinoma (MCC)/neuroendocrine carcinoma of the skin and the clinicopathological similarities between SCLC acid MCC, we wished to determine if this was also the case in MCC, Twenty-nine MCC specimens from 23 patients were examined for deletions at 10 loci on 9p and I on 9p. No loss of heterozygosity (LO H) was peen in 9 patients including 2 for which tumour and cell line DNAs were examined. Four patients had LOH for all informative loci on 9p, Ten tumours showed more limited regions of loss on 9p, and from these 2 common regions of deletion were determined, Half of all informative cases had LOH at D95168, the most telomeric marker examined, and 3 specimens showed loss of only D9S168, A second region (InFNA-D9S126) showed L0H in 10(44%) cases, and case MCC26 showed LOH for only D9S126, implicating genes centromeric of the CDKN2A locus. No mutations in the coding regions of p16 were seen in 7 cell lines tested, and reactivity to anti-p16 antibody was seen in all Il tumour specimens examined and in 6 of 7 cell lines from 6 patients. Furthermore, all cell lines examined reacted with anti-p 14' antibody, These results suggest that neither transcript of the CDKN2A locus is the target of deletions on 9p in MCC and imply the existence of tumour-suppressor genes mapping both centromeric and telomeric of this locus. (C) 2001 Wiley-Liss, Inc.

Similar Markov chains

enin et al. (2000) recently introduced the idea of similarity in the context of birth-death processes. This paper examines the extent to which their results can be extended to arbitrary Markov chains. It is proved that, under a variety of conditions, similar chains are strongly similar in a sense which is described, and it is shown that minimal chains are strongly similar if and only if the corresponding transition-rate matrices are strongly similar. A general framework is given for constructing families of strongly similar chains; it permits the construction of all such chains in the irreducible case.

Supervisory control of wastewater treatment plants by combining principal component analysis and fuzzy c-means clustering

In this paper a methodology for integrated multivariate monitoring and control of biological wastewater treatment plants during extreme events is presented. To monitor the process, on-line dynamic principal component analysis (PCA) is performed on the process data to extract the principal components that represent the underlying mechanisms of the process. Fuzzy c-means (FCM) clustering is used to classify the operational state. Performing clustering on scores from PCA solves computational problems as well as increases robustness due to noise attenuation. The class-membership information from FCM is used to derive adequate control set points for the local control loops. The methodology is illustrated by a simulation study of a biological wastewater treatment plant, on which disturbances of various types are imposed. The results show that the methodology can be used to determine and co-ordinate control actions in order to shift the control objective and improve the effluent quality.

Path integrals for continuous-time Markov chains

This note presents a method of evaluating the distribution of a path integral for Markov chains on a countable state space.

Set-valued Markov chains and negative semitrajectories of discretized dynamical systems

Computer simulation of dynamical systems involves a phase space which is the finite set of machine arithmetic. Rounding state values of the continuous system to this grid yields a spatially discrete dynamical system, often with different dynamical behaviour. Discretization of an invertible smooth system gives a system with set-valued negative semitrajectories. As the grid is refined, asymptotic behaviour of the semitrajectories follows probabilistic laws which correspond to a set-valued Markov chain, whose transition probabilities can be explicitly calculated. The results are illustrated for two-dimensional dynamical systems obtained by discretization of fractional linear transformations of the unit disc in the complex plane.

Integrals for continuous-time Markov chains

This paper presents a method of evaluating the expected value of a path integral for a general Markov chain on a countable state space. We illustrate the method with reference to several models, including birth-death processes and the birth, death and catastrophe process. (C) 2002 Elsevier Science Inc. All rights reserved.

Adaptive multiscale principal components analysis for online monitoring of wastewater treatment

Fault detection and isolation (FDI) are important steps in the monitoring and supervision of industrial processes. Biological wastewater treatment (WWT) plants are difficult to model, and hence to monitor, because of the complexity of the biological reactions and because plant influent and disturbances are highly variable and/or unmeasured. Multivariate statistical models have been developed for a wide variety of situations over the past few decades, proving successful in many applications. In this paper we develop a new monitoring algorithm based on Principal Components Analysis (PCA). It can be seen equivalently as making Multiscale PCA (MSPCA) adaptive, or as a multiscale decomposition of adaptive PCA. Adaptive Multiscale PCA (AdMSPCA) exploits the changing multivariate relationships between variables at different time-scales. Adaptation of scale PCA models over time permits them to follow the evolution of the process, inputs or disturbances. Performance of AdMSPCA and adaptive PCA on a real WWT data set is compared and contrasted. The most significant difference observed was the ability of AdMSPCA to adapt to a much wider range of changes. This was mainly due to the flexibility afforded by allowing each scale model to adapt whenever it did not signal an abnormal event at that scale. Relative detection speeds were examined only summarily, but seemed to depend on the characteristics of the faults/disturbances. The results of the algorithms were similar for sudden changes, but AdMSPCA appeared more sensitive to slower changes.