Use this URL to cite or link to this record in EThOS:
Title: Essays in econometrics
Author: Sher, Galen
ISNI:       0000 0004 7234 1090
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Chapter 1 I revisit claims in the literature about the relationship between financial development and economic growth. I find that the Solow-Swan growth determinants are important omitted variables in the existing empirical literature. The role for credit in determining growth is attenuated once one controls for investment. The positive relationship between growth and credit in the literature is explained by a combination of the well known positive relationship between growth and investment and a positive relationship between investment and credit. However, there is evidence for a negative relationship between growth and credit at levels of credit above 90% of output and this relationship is not explained by any declining effect of credit on investment at these high levels of credit. Chapter 2 I present a model for describing the dependence of financial intermediary net worth on the term to maturity structure of nominal interest rates. Such a model can be used to predict the change in net worth caused by a change in nominal interest rates. I show how to quantify the uncertainty in this change in net worth that arises from term structure parameter estimation and measurement error in accounting data. I apply the methods to measure the exposure of US banks to changes in nominal interest rates. Chapter 3 I consider a measure of serial dependence in stochastic processes based on the compressibility of samples under modern compression algorithms. I relate the measure, which depends on the entropy rate of the stochastic process, to traditional measures of dependence, including those based on entropy like Hong and White [2004]. I present numerical evidence on the finite and large sample distributions of the measure under serial independence, that can be used to choose critical values for hypothesis tests for dependence.
Supervisor: Bond, Steve Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available