Essays on financial time-series analysis and applications

Abstract

This dissertation contains three essays about financial time series and applications. Specifically, we model and predict financial volatility and investigate the emission trading system. In the first study, we investigate the economic and financial drivers of volatility changes and integrate them into stock market volatility forecasting. We first collect a diverse set of predictor variables and analyze them within a unified framework. We discover that only a small number of variables contain significant predictive information, and that the Chinese stock market return significantly predicts U.S. stock market volatility. Using the HAR-LASSO procedure, we integrate the drivers’ predictive information and forecast short-term, medium-term, and long-term market volatilities. Through various volatility timing strategies, we verify that HAR-LASSO-based portfolios lead to outstanding investment performance, regardless of the strategies and forecasting horizons. These results not only economically justify the procedure, but also provide meaningful financial implications of accurate volatility forecasting. Recently, to account for low-frequency market dynamics, several volatility models, employing high-frequency financial data, have been developed. However, in financial markets, we often observe that financial volatility processes depend on economic states, so they have a state heterogeneous structure. In the second study, to investigate state heterogeneous market dynamics based on high-frequency data, we introduce a novel volatility model based on a continuous Ito diffusion process whose intraday instantaneous volatility process evolves depending on the exogenous state variable, as well as its integrated volatility. We call it the state heterogeneous GARCH-Ito (SG-Ito) model. We suggest a quasi-likelihood estimation procedure with the realized volatility proxy and establish its asymptotic behaviors. Moreover, to test the low-frequency state heterogeneity, we develop a Wald test-type hypothesis testing procedure. The results of empirical studies suggest the existence of leverage, investor attention, market illiquidity, stock market comovement, and post-holiday effect in S&P 500 index volatility. In the third study, we analyze the relationship between the price of carbon-intensive fuel and the stock prices of renewable energy companies, incorporating the price of carbon in the European Union emission trading system (EU ETS). Specifically, we employ wavelet methods to reconstruct time series with specific levels of persistence, reducing noise, trend, and seasonal components. Using these wavelet-adjusted series, we conduct a regression analysis that considers exogenous factors that may influence the demand for electricity and emission allowances. Subsequently, we estimate vector autoregressive models and obtain a connectedness measure and impulse response functions. The results consistently imply that increases in coal prices have (counterintuitively) a negative effect on renewable energy stock prices. Moreover, we show that this can be explained by a negative relationship between coal and carbon prices and a positive relationship between carbon prices and renewable energy stock prices. This study contributes to the literature by uncovering the negative relationship between the price of carbon-intensive fuel and renewable energy stock prices by applying a suitable filtering procedure.