<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-22T14:46:48Z</responseDate><request verb="GetRecord" identifier="oai:uwspace.uwaterloo.ca:10012/7028" metadataPrefix="dim">https://dspacemainprd01.lib.uwaterloo.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:uwspace.uwaterloo.ca:10012/7028</identifier><datestamp>2017-03-21T21:44:08Z</datestamp><setSpec>com_10012_9924</setSpec><setSpec>com_10012_1</setSpec><setSpec>col_10012_9934</setSpec><setSpec>col_10012_6</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:doc="http://www.lyncode.com/xoai" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="author" authority="246b0164-fa39-4d56-8836-3368ea62a88a" confidence="600">Men, Zhongxian</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2012-09-26T17:32:44Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2012-09-26T17:32:44Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2012-09-26T17:32:44Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="submitted">1012</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/10012/7028</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en">Stochastic volatility (SV) models provide a natural framework for a&#xd;
representation of time series for financial asset returns. As a&#xd;
result, they have become increasingly popular in the finance&#xd;
literature, although they have also been applied in other fields&#xd;
such as signal processing, telecommunications, engineering, biology,&#xd;
and other areas.&#xd;
&#xd;
In working with the SV models, an important issue arises as how to&#xd;
estimate their parameters efficiently and to assess how well they&#xd;
fit real data. In the literature, commonly used estimation methods&#xd;
for the SV models include general methods of moments, simulated&#xd;
maximum likelihood methods, quasi Maximum likelihood method, and&#xd;
Markov Chain Monte Carlo (MCMC) methods. Among these approaches,&#xd;
MCMC methods are most flexible in dealing with complicated structure&#xd;
of the models. However, due to the difficulty in the selection of&#xd;
the proposal distribution for Metropolis-Hastings methods, in&#xd;
general they are not easy to implement and in some cases we may also&#xd;
encounter convergence problems in the implementation stage. In the&#xd;
light of these concerns, we propose in this thesis new estimation&#xd;
methods for univariate and multivariate SV models. In the simulation&#xd;
of latent states of the heavy-tailed SV models, we recommend the&#xd;
slice sampler algorithm as the main tool to sample the proposal&#xd;
distribution when the Metropolis-Hastings method is applied. For the&#xd;
SV models without heavy tails, a simple Metropolis-Hastings method&#xd;
is developed for simulating the latent states. Since the slice&#xd;
sampler can adapt to the analytical structure of the underlying&#xd;
density, it is more efficient. A sample point can be obtained from&#xd;
the target distribution with a few iterations of the sampler,&#xd;
whereas in the original Metropolis-Hastings method many sampled&#xd;
values often need to be discarded.&#xd;
&#xd;
In the analysis of multivariate time series, multivariate SV models&#xd;
with more general specifications have been proposed to capture the&#xd;
correlations between the innovations of the asset returns and those&#xd;
of the latent volatility processes. Due to some restrictions on the&#xd;
variance-covariance matrix of the innovation vectors, the estimation&#xd;
of the multivariate SV (MSV) model is challenging. To tackle this&#xd;
issue, for a very general setting of a MSV model we propose a&#xd;
straightforward MCMC method in which a Metropolis-Hastings method is&#xd;
employed to sample the constrained variance-covariance matrix, where&#xd;
the proposal distribution is an inverse Wishart distribution. Again,&#xd;
the log volatilities of the asset returns can then be simulated via&#xd;
a single-move slice sampler.&#xd;
&#xd;
Recently, factor SV models have been proposed to extract hidden&#xd;
market changes. Geweke and Zhou (1996) propose a factor SV model&#xd;
based on factor analysis to measure pricing errors in the context of&#xd;
the arbitrage pricing theory by letting the factors follow the&#xd;
univariate standard normal distribution. Some modification of this&#xd;
model have been proposed, among others, by Pitt and Shephard (1999a)&#xd;
and  Jacquier et al. (1999). The main feature of the factor SV&#xd;
models is that the factors follow a univariate SV process, where the&#xd;
loading matrix is a lower triangular matrix with unit entries on the&#xd;
main diagonal. Although the factor SV models have been successful in&#xd;
practice, it has been recognized that the order of the component may&#xd;
affect the sample likelihood and the selection of the factors.&#xd;
Therefore, in applications, the component order has to be considered&#xd;
carefully. For instance, the factor SV model should be fitted to&#xd;
several permutated data to check whether the ordering affects the&#xd;
estimation results. In the thesis, a new factor SV model is&#xd;
proposed. Instead of setting the loading matrix to be lower&#xd;
triangular, we set it to be column-orthogonal and assume that each&#xd;
column has unit length. Our method removes the permutation problem,&#xd;
since when the order is changed then the model does not need to be&#xd;
refitted. Since a strong assumption is imposed on the loading&#xd;
matrix, the estimation seems even harder than the previous factor&#xd;
models. For example, we have to sample columns of the loading matrix&#xd;
while keeping them to be orthonormal. To tackle this issue, we use&#xd;
the Metropolis-Hastings method to sample the loading matrix one&#xd;
column at a time, while the orthonormality between the columns is&#xd;
maintained using the technique proposed by Hoff (2007). A von&#xd;
Mises-Fisher distribution is sampled and the generated vector is&#xd;
accepted through the Metropolis-Hastings algorithm.&#xd;
&#xd;
Simulation studies and applications to real data are conducted to&#xd;
examine our inference methods and test the fit of our model.&#xd;
Empirical evidence illustrates that our slice sampler within MCMC&#xd;
methods works well in terms of parameter estimation and volatility&#xd;
forecast. Examples using financial asset return data are provided to&#xd;
demonstrate that the proposed factor SV model is able to&#xd;
characterize the hidden market factors that mainly govern the&#xd;
financial time series. The Kolmogorov-Smirnov  tests conducted on&#xd;
the estimated models indicate that the models do a reasonable job in&#xd;
terms of describing real data.</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en">en</dim:field>
   <dim:field mdschema="dc" element="publisher" lang="en">University of Waterloo</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Bayesian inference</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Stochastic volatility</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Markov Chain Monte Carlo</dim:field>
   <dim:field mdschema="dc" element="subject" qualifier="program" lang="en">Statistics</dim:field>
   <dim:field mdschema="dc" element="title" lang="en">Bayesian Inference for Stochastic Volatility Models</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Doctoral Thesis</dim:field>
   <dim:field mdschema="dc" element="pending" lang="en">false</dim:field>
   <dim:field mdschema="uws-etd" element="degree" qualifier="department" lang="en">Statistics and Actuarial Science</dim:field>
   <dim:field mdschema="uws-etd" element="degree" lang="en">Doctor of Philosophy</dim:field>
   <dim:field mdschema="uws" element="typeOfResource" lang="en">Text</dim:field>
   <dim:field mdschema="uws" element="peerReviewStatus" lang="en">Unreviewed</dim:field>
   <dim:field mdschema="uws" element="scholarLevel" lang="en">Graduate</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim>
</metadata></record></GetRecord></OAI-PMH>