Estimating parameters in the presence of many nuisance parameters

Wu, Billy (2013) Estimating parameters in the presence of many nuisance parameters. MPhil thesis, London School of Economics and Political Science.

Preview

PDF Download (406kB) | Preview

Abstract

This paper considers estimation of parameters for high-dimensional time series with the presence of many nuisance parameters. In particular we are interested in data consisting of p time series of length n, with p to be as large or even larger than n. Here we consider the composite-likelihood estimation and the profile quasi-likelihood estimation. The asymptotic properties of these methodologies are investigated. Simulations are used to illustrate our both of these methods and explore the performance of these methods.

Item Type:	Thesis (MPhil)
Additional Information:	© 2013 Billy Wu
Uncontrolled Keywords:	composite likelihood, nuisance parameter, profile likelihood, quasi-likelihood, root-n convergence, time series.
Library of Congress subject classification:	H Social Sciences > HA Statistics
Sets:	Departments > Statistics
Supervisor:	Yao, Qiwei
URI:	http://etheses.lse.ac.uk/id/eprint/545

Actions (login required)

Record administration - authorised staff only