Spelling suggestions: "subject:"stochastic"" "subject:"tochastic""
1 |
A stochastic realization and model reduction approach to streamflow modelingRamos, José A 12 1900 (has links)
No description available.
|
2 |
THE CONTROL OF NONLINEAR STOCHASTIC CONTROL SYSTEMS UNDER DISCOUNTED PERFORMANCE CRITERIAHarris, Cliff Andrew, 1942- January 1970 (has links)
No description available.
|
3 |
Stochastic models for asset and liability modelling in South Africa or elsewhereMaitland, Alexander James 16 September 2011 (has links)
Ph. D, Faculty of Science, University of Witwatersrand, 2011 / Research in the area of stochastic models for actuarial use in South Africa is limited to
relatively few publications. Until recently, there has been little focus on actuarial
stochastic models that describe the empirical stochastic behaviour of South African
financial and economic variables. A notable exception is Thomson’s (1996) proposed
methodology and model. This thesis presents a collection of five papers that were
presented at conferences or submitted for peer review in the South African Actuarial
Journal between 1996 and 2006. References to subsequent publications in the field are
also provided. Such research has implications for medium and long-term financial
simulations, capital adequacy, resilience reserving and asset allocation benchmarks as
well as for the immunization of short-term interest rate risk, for investment policy
determination and the general quantification and management of risk pertaining to those
assets and liabilities.
This thesis reviews Thomson’s model and methodology from both a statistical and
economic perspective, and identifies various problems and limitations in that approach.
New stochastic models for actuarial use in South Africa are proposed that improve the
asset and liability modelling process and risk quantification. In particular, a new Multiple
Markov-Switching (MMS) model framework is presented for modelling South African
assets and liabilities, together with an optimal immunization framework for nominal
liability cash flows. The MMS model is a descriptive model with structural features and
parameter estimates based on historical data. However, it also incorporates theoretical
aspects in its design, thereby providing a balance between purely theoretical models and those based only on empirical considerations.
|
4 |
The behaviour of stochastic rumours.Belen, Selma January 2008 (has links)
This thesis presents results concerning the limiting behaviour of stochastic rumour processes. The first result involves our published analysis of the evolution for the general initial conditions of the (common) deterministic limiting version of the classical Daley-Kendall and Maki-Thompson stochastic rumour models, [14]. The second result being also part of the general analysis in [14] involves a new approach to stiflers in the rumour process. This approach aims at distinguishing two main types of stiflers. The analytical and stochastic numerical results of two types of stiflers in [14] are presented in this thesis. The third result is that the formulae to find the total number of transitions of a stochastic rumour process with a general case of the Daley-Kendall and Maki-Thompson classical models are developed and presented here, as already presented in [16]. The fourth result is that the problem is taken into account as an optimal control problem and an impulsive control element is introduced to minimize the number of final ignorants in the stochastic rumour process by repeating the process. Our published results are presented in this thesis as appeared in [15] and [86]. Numerical results produced by our algorithm developed for the extended [MT] model and [DK] model are demonstrated by tables in all details of numerical values in the appendices. / Thesis (Ph.D.) - University of Adelaide, School of Mathematical Sciences, 2008
|
5 |
Multi-objective stochastic path planningDasgupta, Sumantra 15 May 2009 (has links)
The present research formulates the path planning as an optimization problem
with multiple objectives and stochastic edge parameters. The first section introduces
different variants of the PP problem and discusses existing solutions to the problem. The
next section introduces and solves various versions of the PP model within the scope of
this research. The first three versions describe a single entity traveling from a single
source to a single destination node. In the first version, the entity has a single objective
and abides by multiple constraints. The second version deals with an entity traveling
with multiple objectives and multiple constraints. The third version is a modification of
the second version where the actual probability distributions of travel times along edges
are known. The fourth and final version deals with multiple heterogeneous entities
routed from multiple sources (supply nodes) to multiple destinations (demand nodes)
along capacitated edges. Each of these formulations is solved by using either exact
algorithms or heuristics developed in this research. The performance of each
algorithm/heuristic is discussed in the final section. The main contributions of this
research are: 1. Provide a framework for analyzing PP in presence of multiple objectives and
stochastic edge parameters.
2. Identify candidate constraints where clustering based multi-level programming
can be applied to eliminate infeasible edges.
3. Provide an exact O (V.E) algorithm for building redundant shortest paths.
4. Provide an O (V.E+C2) heuristic for generating Pareto optimal shortest paths in
presence of multiple objectives where C is the upper bound for path length. The
complexity can be further reduced to O (V.E) by using graphical read-out of the
Pareto frontier.
5. Provide a cost structure which can capture multiple key probability distribution
parameters of edge variables. This is in contrast with usual techniques which just
capture single parameters like the mean or the variance of distributions.
6. Provide a MIP formulation to a multi-commodity transportation problem with
multiple decision variables, stochastic demands and uncertain edge/route
capacities.
7. Provide an alternate formulation to the classic binary facility selection problem.
|
6 |
Stochastic resonance in nanoscale systemsSaha, Aditya Unknown Date
No description available.
|
7 |
STOCHASTIC ADC WITH RANDOM U-QUADRATIC DISTRIBUTED REFERENCE VOLTAGES TO UNIFORMLY DISTRIBUTE COMPARATORS TRIP POINTSCeekala, Mithun 23 April 2013 (has links)
This thesis presents a new architecture of stochastic Analog-to-Digital converter (ADC). A standard Stochastic ADC uses comparator random offset as the trip point while all the comparators have the same reference voltages. Since the offset of a basic comparator depends on a number of independent random variables, the offset will follow randomly distributed Gaussian function. The input dynamic range of this standard stochastic ADC is ±?. For 90nm technology ? value is around 153mV. A technique is presented that converts overall transfer function of a stochastic ADC i.e. Gaussian distribution into almost uniformly distribution with a wider range. With the proposed technique, an input dynamic range of ± 153mV and ENOB of 4bits of standard stochastic ADC are increased to variable input dynamic range of ±250mV to ±500mV and ENOB of 6bits.
|
8 |
Stochastic Approximation and Its Application in MCMCCheng, Yichen 16 December 2013 (has links)
Stochastic approximation has been widely used since first proposed by Herbert Robbins and Sutton Monro in 1951. It is an iterative stochastic method that attempts to find the zeros of functions that cannot be computed directly. In this thesis, we used the technique in several different aspects. It was used in the analysis of large geostatistical data, in the improvement of simulated annealing algorithm also, as well as for NMR protein structure determination.
1. We proposed a resampling based Stochastic approximation method for the analysis of large geostatistical data. The main difficulty that lies in the analysis of geostatistical data is the computation time is extremely long when the sample size becomes large. Our proposed method only use a small portion of the data at each iteration. Each time, we update our estimators based on a randomly selected subset of the data using stochastic approximation. In this way, we use the information from the whole data set while keep the computation time almost irrelevant to the sample size. We proved the consistency of our estimator and showed by simulation study that the computation time is much reduced compared to other existing methods.
2. Simulated Annealing algorithm has been widely used for optimization problems. However, it can not guarantee the global optima to be located unless a logarithmic cooling schedule is used. However, the logarithm rate is so slow that no one can afford such a long cpu time. We proposed a new stochastic optimization algorithm, the so-called simulated stochastic approximation annealing (SAA) algorithm, which is a combination of simulated annealing and the stochastic approximation Monte Carlo (SAMC) algorithm. It is shown that the new algorithm can work with a cooling schedule that decreases much faster than in the logarithmic cooling schedule while guarantee the global optima to be reached when temperature tends to zero.
3. Protein Structure determination is a very important topic in computational biology. It aims to determine different conformations for each protein, which helps to understand biological functions such as protein-protein interactions, protein-DNA interactions and so on. Protein structure determination consists of a series of steps and peak picking is a very important step. It is the prerequisite for all other steps. Manually pick the peaks is very time consuming. To automate this process, several methods have been proposed. However, due to the complexity of NMR spectra, the existing method is hard to distinguish false peaks and true peaks perfectly. The main difficulty lies in identifying true peaks with low intensity and overlapping peaks.
We propose to model the spectrum as a mixture of bivariate Gaussian densities and used stochastic approximation Monte Carlo (SAMC) method as the computational approach to solve this problem. Essentially, by putting the peak picking problem into a Bayesian framework, we turned it into a model selection problem. Because Bayesian method will automatically penalize including too much component into the model, our model will distinguish true peaks from noises without pre-process of the data.
|
9 |
Some results on pinching matricesKo, Chiu-chan., 高超塵. January 2003 (has links)
published_or_final_version / abstract / toc / Mathematics / Master / Master of Philosophy
|
10 |
Aspects of modelling stochastic volatilityTsang, Wai-yin, 曾慧賢 January 2000 (has links)
published_or_final_version / Statistics and Actuarial Science / Master / Master of Philosophy
|
Page generated in 0.1249 seconds