The real effects of banking supervision:evidence from enforcement actions

We present a novel way to examine macro-financial linkages by focusing on the real effects of bank supervisors’ enforcement actions. Exploiting plausibly exogenous variation in supervisory monitoring intensity, we show that enforcement actions in single-market banks trigger temporarily large adverse effects for the macroeconomy by reducing personal income growth, the number of establishments, and increasing unemployment. These effects are related to contractions in bank lending and liquidity creation, and are more pronounced when we consider enforcement actions on both single-market and multi-market banks, and in counties with fewer banks and greater external financial dependence.

On the solution of a class of capital investment problems

In this work the solution of a class of capital investment problems is considered within the framework of mathematical programming. Upon the basis of the net present value criterion, the problems in question are mainly characterized by the fact that the cost of capital is defined as a non-decreasing function of the investment requirements. Capital rationing and some cases of technological dependence are also included, this approach leading to zero-one non-linear programming problems, for which specifically designed solution procedures supported by a general branch and bound development are presented. In the context of both this development and the relevant mathematical properties of the previously mentioned zero-one programs, a generalized zero-one model is also discussed. Finally,a variant of the scheme, connected with the search sequencing of optimal solutions, is presented as an alternative in which reduced storage limitations are encountered.

Studies on the generalisation of Gaussian processes and Bayesian neural networks

The assessment of the reliability of systems which learn from data is a key issue to investigate thoroughly before the actual application of information processing techniques to real-world problems. Over the recent years Gaussian processes and Bayesian neural networks have come to the fore and in this thesis their generalisation capabilities are analysed from theoretical and empirical perspectives. Upper and lower bounds on the learning curve of Gaussian processes are investigated in order to estimate the amount of data required to guarantee a certain level of generalisation performance. In this thesis we analyse the effects on the bounds and the learning curve induced by the smoothness of stochastic processes described by four different covariance functions. We also explain the early, linearly-decreasing behaviour of the curves and we investigate the asymptotic behaviour of the upper bounds. The effect of the noise and the characteristic lengthscale of the stochastic process on the tightness of the bounds are also discussed. The analysis is supported by several numerical simulations. The generalisation error of a Gaussian process is affected by the dimension of the input vector and may be decreased by input-variable reduction techniques. In conventional approaches to Gaussian process regression, the positive definite matrix estimating the distance between input points is often taken diagonal. In this thesis we show that a general distance matrix is able to estimate the effective dimensionality of the regression problem as well as to discover the linear transformation from the manifest variables to the hidden-feature space, with a significant reduction of the input dimension. Numerical simulations confirm the significant superiority of the general distance matrix with respect to the diagonal one.In the thesis we also present an empirical investigation of the generalisation errors of neural networks trained by two Bayesian algorithms, the Markov Chain Monte Carlo method and the evidence framework; the neural networks have been trained on the task of labelling segmented outdoor images.

Apoptotic and necrotic effects of hexanedione derivatives on the human neuroblastoma line SK-N-SH

The potential cytotoxicity of two hexanedione food additives (2,3 and 3,4 isomers) was evaluated in comparison with the neurotoxic hexane metabolite 2,5-hexanedione in the human SK-N-SH neuroblastoma line using the MTT assay to indicate mitochondrial dehydrogenase activity and flow cytometry to monitor the cell cycle over 48 h. The IC50s of the 2,3-hexanedione (3.3 ± 0.1 mM) and 3,4-hexanedione (3.5 ± 0.1 mM), indicated that the sensitivity of the cells was approximately seven-fold greater to these toxins compared with the 2,5 derivative (IC50 of 22.4 ± 0.2 mM). Comparison between the respective IC50s of the 2,3-hexanedione and 3,4-hexanedione revealed no difference between the two isomers in terms of their effects on MTT turnover. With flow cytometry analysis, all three hexanediones showed increases in apoptosis within their respective concentration ranges of toxicity shown previously by MTT. In the presence of 2,5-hexanedione, between 8.5 and 17 mM concentrations, there was a significant increase in apoptotic nucleoids which was accompanied by a significant fall in the percentage of nucleoids in the G0/G1 phase (72.4 ± 0.3-45.3 ± 0.6%,), and a rise in the numbers of cells in the G2/M phase. This is likely to indicate growth arrest at cell cycle G2/M checkpoint in response to toxin damage. G2/M accumulation was also shown with 3,4 and 2,3 HD, which was maximal at much lower concentrations (approximately 4 and 3 mM, respectively). Arrest at G1 and G2/M phase is indicative of inhibition of the cell cycle at the stages of DNA replication and chromosome segregation, respectively. It was also apparent that flow cytometry, rather than the MTT assay, did distinguish between the effects of the α-diketones 2,3-hexanedione and 3,4-hexanedione on the cell cycle. At a concentration of 5.8 mM 3,4-hexanedione, the percentage of apoptotic nucleoids was 10.9 ± 0.8% whilst apoptosis induced by 3,4-hexanedione had already reached a maximal level of 60.4 ± 0.5%. In summary, flow cytometry indicated that the 3,4-hexanedione derivative was more toxic than its 2,3 isomer and that both food additives caused interruption in the neuroblastoma cell cycle and further investigation may be required to assess if these α-diketones present in diets pose any possible risks to human health. © 2006 Elsevier Ireland Ltd. All rights reserved.

On the theory of autosoliton propagation in optical fibers guided by in-line nonlinear devices

A theoretical model is developed to describe the propagation of ultra-short optical pulses in fiber transmission systems in the quasi-linear regime, with periodically inserted in-line lumped nonlinear optical devices. Stable autosoliton solutions are obtained for a particular application of the general theory.

On the numerical solution of a Cauchy problem in an elastostatic half-plane with a bounded inclusion

We propose an iterative procedure for the inverse problem of determining the displacement vector on the boundary of a bounded planar inclusion given the displacement and stress fields on an infinite (planar) line-segment. At each iteration step mixed boundary value problems in an elastostatic half-plane containing the bounded inclusion are solved. For efficient numerical implementation of the procedure these mixed problems are reduced to integral equations over the bounded inclusion. Well-posedness and numerical solution of these boundary integral equations are presented, and a proof of convergence of the procedure for the inverse problem to the original solution is given. Numerical investigations are presented both for the direct and inverse problems, and these results show in particular that the displacement vector on the boundary of the inclusion can be found in an accurate and stable way with small computational cost.

A comparison of the apoptotic and cytotoxic effects of hexanedione derivatives on human non-neuronal lines and the neuroblastoma line SH-SY5Y

The effects of the alpha-diketone derivatives 2,3- and 3,4-hexanediones were investigated in three non-neuronal cell lines (MCF7, HepG2 and CaCo-2) as well as in the neuroblastoma line, SH-SY5Y. The MTT reduction assay was employed to determine the necrotic effects of the alpha-diketones and the neurotoxin 2,5-hexanedione over 4, 24 and 48 hr exposures. Flow cytometry was also used to study the effects of the three isomers on the cell cycle of the SH-SY5Y line only. With 2,5-hexanedione, the mean MTT IC50 decreased more than 10-fold from 4 to 48 hr. The toxicities of both alpha-diketones were similar, with a more than 18-fold increase in sensitivity of the SH-SY5Y at 24 hr compared to that of 4 hr. With flow cytometry at 48 hr, SH-SY5Y apoptosis with 2,5-hexanedione rose throughout the concentration range evaluated (0-30 mM) while 2,3- and 3,4-hexanediones showed apoptosis over the concentration range 1-1.6 mM, with 3,4-hexanedione being the more potent compared to the 2,3-isomer. At 1.6 mM nearly all the cells had entered apoptosis in the presence of the 3,4-isomer, (94.9 ± 1.4%) but only 57.5 ±4.1% of the 2,3-isomer-treated cells had reached that stage. The 2,3-and 3,4-isomers in diets alone may not pose a serious threat to human health. Further studies may be necessary to evaluate the effects of other dietary components on their toxicity. These alpha-diketones also display a degree of toxic selectivity towards neuroblastoma cells, which may have therapeutic implications.

On the theory of autosoliton propagation in optical fibers guided by in-line nonlinear devices

A theoretical model is developed to describe the propagation of ultra-short optical pulses in fiber transmission systems in the quasi-linear regime, with periodically inserted in-line lumped nonlinear optical devices. Stable autosoliton solutions are obtained for a particular application of the general theory.

Impact of optical phase conjugation on the nonlinear Shannon limit

Compensation of the detrimental impacts of nonlinearity on long-haul wavelength division multiplexed system performance is discussed, and the difference between transmitter, receiver and in-line compensation analyzed. We demonstrate that ideal compensation of nonlinear noise could result in an increase in the signal-to-noise ratio (measured in dB) of 50%, and that reaches may be more than doubled for higher order modulation formats. The influence of parametric noise amplification is discussed in detail, showing how increased numbers of optical phase conjugators may further increase the received signal-tonoise ratio. Finally the impact of practical real world system imperfections, such as polarization mode dispersion, are outlined.