Electronic Theses and Dissertations

Date of Award


Document Type


Degree Name

Ph.D. in Mathematics



First Advisor

Dao D. Nguyen

Second Advisor

Xin X. Dang

Third Advisor

Hailin H. Sang

Relational Format



In this dissertation, We show the results of our researches in statistical sampling, functional optimization, and methodology for partially observed Markov process (POMP) models. In statistical sampling, we introduce a p-generalized smoothing method that enables the Langevin-Monte Carlo method to generate a sample from a log concave distribution weakly smoothing potential function. For our optimization research, we introduce an accelerated inexact gradient (AIG) method. Combining the strengths while mitigating the weakness of its parent methods: gradient descent and Nesterov's accelerated gradient, AIG converges with excellent rates for both convex and non-convex optimization problems for smooth objective functions. Furthermore, we also adapt AIG into iterated filtering (IF), which is a parameter estimation method on the partially observed Markov process models. This vastly improves both the accuracy and precision of IF, allowing it to perform favorably compared to the state of the art methods in our simulation example. Finally, we introduce our R package "is2", which extends some useful Monte Carlo statistical methodologies to improve on convergence rates and inherits the flexible framework of R package "pomp". A variety of efficient statistical methods for POMP models have been developed including fixed lag smoothing, second-order iterated smoothing, momentum iterated filtering, average iterated filtering, accelerated iterated filtering and particle iterated filtering. In this dissertation, we show the utility of these methodologies using two toy problems. We also demonstrate the applicability of our methods on a more complex model, employing a nonlinear epidemiological model with a discrete population, seasonality, and extra-demographic stochasticity.

Included in

Mathematics Commons



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.