Essays in econometrics

Abstract

This thesis consists of two main parts. The first part deals with an analysis of realized volatility and its relationship with market microstructure problem. The second part of the thesis presents a time trend analysis in a panel data framework, with a semiparametric approach. Chapter 1 introduces the topics that I embark upon the thesis. In particular, I motivate the interest in realized volatility and market microstructure problem in the first part of the thesis, with a factor model approach. Then, in the second part, the motivation is on the estimation of time varying coefficient trend functions in a panel data case, using nonparametric estimation methods. Chapter 2 proposes a literature review on realized volatility and factor models, while focusing on the seminal papers and models that the theoretical literature suggests and also provides the empirical evidence observed in financial markets. Chapter 3 develops a theoretical model to forecast the realized volatility consistently and efficiently for large dimensional datasets and also addresses the solution for noise problem coming out of volatility estimation in the presence of market microstructure effects. Chapter 4 provides the empirical analysis and results on a sample of S&P 500 stocks following the methodology and models suggested in Chapter 3. Chapter 5 focuses on developing a semiparametric panel model to explain the time trend function. Profile likelihood estimators (PLE) are proposed and their statistical properties are studied. We apply our methods to the UK regional temperatures. Finally, forecasting based on the proposed model is studied. Chapter 6 concludes, summarizing the main results and contributions of the thesis.