No description available.

No description available.

No description available.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第20517号 / 情博第645号 / 新制||情||111(附属図書館) / 京都大学大学院情報学研究科システム科学専攻 / (主査)教授 髙橋 豊, 教授 太田 快人, 教授 大塚 敏之, 准教授 増山 博之 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Šiame darbe nagrinėta tam tikra asimptotiškai homogeninė Markovo grandinė ir rasta jos stacionaraus skirstinio uodegos asimptotika. Nagrinėta grandinė negali būti ištirta šiuo metu žinomais metodais, todėl darbas turi praktinę reikšmę. Spręstas uždavinys aktualus sunkių uodegų analizėje. / In this work we have investigated some asymptotically homogeneous Markov chain and found asymptotics of the stationary distribution tail. To our best knowledge, considered chain cannot be investigated by means of existing methods, hence obtained results have practical value. Solved problem is relevant in heavy tail analysis.

In this thesis, we derive, extend and generalise various aspects of impact theory and splash dynamics. Our methods throughout will involve isolating small parameters in our models, which we can utilise using the language of matched asymptotics. In Chapter 1 we briefly motivate the field of impact theory and outline the structure of the thesis. In Chapter 2, we give a detailed review of classical small-deadrise water entry, Wagner theory, in both two and three dimensions, highlighting the key results that we will use in our extensions of the theory. We study oblique water entry in Chapter 3, in which we use a novel transformation to relate an oblique impact with its normal-impact counterpart. This allows us to derive a wide range of solutions to both two- and three-dimensional oblique impacts, as well as discuss the limitations and breakdown of Wagner theory. We return to vertical water-entry in Chapter 4, but introduce the air layer trapped between the impacting body and the liquid it is entering. We extend the classical theory to include this air layer and in the limit in which the density ratio between the air and liquid is sufficiently small, we derive the first-order correction to the Wagner solution due to the presence of the surrounding air. The model is presented in both two dimensions and axisymmetric geometries. In Chapter 5 we move away from Wagner theory and systematically derive a series of splash jet models in order to find possible mechanisms for phenomena seen in droplet impact and droplet spreading experiments. Our canonical model is a thin jet of liquid shot over a substrate with a thin air layer trapped between the jet and the substrate. We consider a variety of parameter regimes and investigate the stability of the jet in each regime. We then use this model as part of a growing-jet problem, in which we attempt to include effects due to the jet tip. In the final chapter we summarise the main results of the thesis and outline directions for future work.

Enumeration of pattern avoiding compositions of numbers Abstract The aim of this work was to find some new results for 3-regular compositions, i.e., for those compositions which avoid the set of patterns {121, 212, 11}. Those compositions can be regarded as a generalization of Carlitz composition. Based on the generating function of compositions avoiding the set of patterns {121, 11} and {212, 11} we derive an upper bound for the coefficients of the power series of the generating function of 3-regular compositions. Using the theory of finite automata we derive its lower bound. We develop this result further by defining 3-block compositions. For the generating function of 3-regular compositions we prove a recursive ralation. Besides that we also compute the generating function of compositions avoiding the set of patterns {312, 321} whose parts are in the set [d]. In the last section we prove that the generating function of Carlitz compositions is transcendental.

Student Number : 9606114D PhD Thesis School of Mathematics Faculty of Science / In this thesis, our primary interest lies in the investigation of the location of the zeros and the asymptotic zero distribution of hypergeometric polynomials. The location of the zeros and the asymptotic zero distribution of general hy- pergeometric polynomials are linked with those of the classical orthogonal polynomials in some cases, notably 2F1 and 1F1 hypergeometric polynomials which have been extensively studied. In the case of 3F2 polynomials, less is known about their properties, including the location of their zeros, because there is, in general, no direct link with orthogonal polynomials. Our intro- duction in Chapter 1 outlines known results in this area and we also review recent papers dealing with the location of the zeros of 2F1 and 1F1 hyperge- ometric polynomials. In Chapter 2, we consider two classes of 3F2 hypergeometric polynomials, each of which has a representation in terms of 2F1 polynomials. Our first result proves that the class of polynomials 3F2(−n, a, b; a−1, d; x), a, b, d 2 R, n 2 N is quasi-orthogonal of order 1 on an interval that varies with the values of the real parameters b and d. We deduce the location of (n−1) of its zeros and dis- cuss the apparent role played by the parameter a with regard to the location of the one remaining zero of this class of polynomials. We also prove re- sults on the location of the zeros of the classes 3F2(−n, b, b−n 2 ; b−n, b−n−1 2 ; x), b 2 R, n 2 N and 3F2 (−n, b, b−n 2 + 1; b − n, b−n+1 2 ; x), n 2 N, b 2 R by using the orthogonality and quasi-orthogonality of factors involved in its representation. We use Mathematica to plot the zeros of these 3F2 hypergeometric polynomials for different values of n as well as for different ranges of the pa- rameters. The numerical data is consistent with the results we have proved. The Euler integral representation of the 2F1 Gauss hypergeometric function is well known and plays a prominent role in the derivation of transformation identities and in the evaluation of 2F1(a, b; c; 1), among other applications (cf. [1], p.65). The general p+kFq+k hypergeometric function has an integral repre- sentation (cf. [37], Theorem 38) where the integrand involves pFq. In Chapter 3, we give a simple and direct proof of an Euler integral representation for a special class of q+1Fq functions for q >= 2. The values of certain 3F2 and 4F3 functions at x = 1, some of which can be derived using other methods, are deduced from our integral formula. In Chapter 4, we prove that the zeros of 2F1 (−n, n+1 2 ; n+3 2 ; z) asymptotically approach the section of the lemniscate {z : |z(1 − z)2| = 4 27 ;Re(z) > 1 3} as n ! 1. In recent papers (cf. [31], [32], [34], [35]), Mart´ınez-Finkelshtein and Kuijlaars and their co-authors have used Riemann-Hilbert methods to derive the asymptotic distribution of Jacobi polynomials P(an,bn) n when the limits A = lim n!1 an n and B = lim n!1 Bn n exist and lie in the interior of certain specified regions in the AB-plane. Our result corresponds to one of the transitional or boundary cases for Jacobi polynomials in the Kuijlaars Mart´ınez-Finkelshtein classification.

In this thesis, we aim at a quantitative understanding of extreme risks. We use heavy-tailed distribution functions to model extreme risks, and use various tools, such as copulas and MRV, to model dependence structures. We focus on modeling as well as quantitatively estimating certain measurements of extreme risks. We start with a credit risk management problem. More specifically, we consider a credit portfolio of multiple obligors subject to possible default. We propose a new structural model for the loss given default, which takes into account the severity of default. Then we study the tail behavior of the loss given default under the assumption that the losses of the obligors jointly follow an MRV structure. This structure provides an ideal framework for modeling both heavy tails and asymptotic dependence. Using HRV, we also accommodate the asymptotically independent case. Multivariate models involving Archimedean copulas, mixtures and linear transforms are revisited. We then derive asymptotic estimates for the Value at Risk and Conditional Tail Expectation of the loss given default and compare them with the traditional empirical estimates. Next, we consider an investor who invests in multiple lines of business and study a capital allocation problem. A randomly weighted sum structure is proposed, which can capture both the heavy-tailedness of losses and the dependence among them, while at the same time separates the magnitudes from dependence. To pursue as much generality as possible, we do not impose any requirement on the dependence structure of the random weights. We first study the tail behavior of the total loss and obtain asymptotic formulas under various sets of conditions. Then we derive asymptotic formulas for capital allocation and further refine them to be explicit for some cases. Finally, we conduct extreme risk analysis for an insurer who makes investments. We consider a discrete-time risk model in which the insurer is allowed to invest a proportion of its wealth in a risky stock and keep the rest in a risk-free bond. Assume that the claim amounts within individual periods follow an autoregressive process with heavy-tailed innovations and that the log-returns of the stock follow another autoregressive process, independent of the former one. We derive an asymptotic formula for the finite-time ruin probability and propose a hybrid method, combining simulation with asymptotics, to compute this ruin probability more efficiently. As an application, we consider a portfolio optimization problem in which we determine the proportion invested in the risky stock that maximizes the expected terminal wealth subject to a constraint on the ruin probability.

By the n-th Fibonacci (respectively Lucas) vector of length m, we mean the vector whose components are the n-th through (n+m-1)-st Fibonacci (respectively Lucas) numbers. For arbitrary m, we express the dot product of any two Fibonacci vectors, any two Lucas vectors, and any Fibonacci vector and any Lucas vector in terms of the Fibonacci and Lucas numbers. We use these formulas to deduce a number of identities involving the Fibonacci and Lucas numbers.