PhD THESIS ---
Director: Christian Brownlees
Based on high-frequency price data, this thesis focuses on estimating the realized covariance and the integrated volatility of asset prices, and applying volatility estimation to price jump detection. The first chapter uses the LASSO procedure to regularize some estimators of high dimensional realized covariance matrices. We establish theoretical properties of the regularized estimators that show its estimation precision and the probability that they correctly reveal the network structure of the assets. The second chapter proposes a novel estimator of the integrated volatility which is the quadratic variation of the continuous part in the price process. This estimator is obtained by truncating the two-scales realized variance estimator. We show its consistency in the presence of market microstructure noise and finite or infinite activity jumps in the price process. The third chapter employs this estimator to design a test to explore the existence of price jumps with noisy price data.
Director: Kristoffer Nimark
This thesis consists of three chapters on topics in macroeconomics and finance. In the first chapter, I use texts from corporate filings of US companies to investigate if liquidity shortages that occurred during the late-2000 financial crisis were different from cases that occur during more normal times. In the second chapter, I quantify narrative evidence from corporate filings to construct a novel dataset on the price-setting behavior of companies. I then use this dataset to investigate what factors cause firms to change the prices of their products or prevent them from doing so. In the third chapter, I use a number of high-frequency financial market estimates to identify the monetary policy shock in a non-recursive Factor Augmented Vector Autoregression of monthly frequency.
Kizkitza Biguri Pastor
Director: Hugo Rodriguez Mendizabal and Filippo Ippolito
This dissertation studies how debt structure and risk management decisions affect firms' investment. The first chapter focuses on building the stylized facts on the relation between debt structure, capital structure and investment when firms' have both, secured and unsecured debt available. Results suggests that i) firms with higher creditworthiness tend to borrow more unsecured debt, ii) higher collateral availability may not lead to more investment and iii) more reliance on unsecured debt leads to more investment. The second chapter uses two identification strategies to test the causal effect of the relations derived in chapter one. I test the hypothesis from a balance sheet and credit channel perspective. Results show that the composition of debt structure of firms has real implications. The higher the unsecured debt in debt structure, the more firms can invest. The explanation for this result is that unsecured debt is more cost-effect in terms of spreads and debt covenants. Finally, the last chapter uses a panel of shocks to the cost of hedging to different firms at different points in time to study the relation between hedging and risk. I exploit the introduction and delisting of commodity derivatives by the CME and other exchanges for identification. I find evidence suggesting that cheaper access to hedging instruments reduces the volatility of cashflows and thus, increases firms' investment.
Director: Jordi Galí
The first chapter, I examine both theoretically and empirically how income uncertainty affects the effectiveness of monetary policy. I consider income risk from potential unemployment, and find that monetary policy has a smaller influence on aggregate demand when unemployment risk is high. I build on the fact that saving arising from a precautionary motive has a smaller interest elasticity. As a consequence, aggregate demand reacts less to the interest rate when uncertainty is high. The second chapter links the build-up of financial risk that led to the recent financial crisis to the preceding period of exceptionally low macroeconomic volatility. The degree of stability that a country has enjoyed before 2007 predicts robustly how much it suffered from the crisis, a result that also holds for individual firms. In the final chapter, I connect this period of low volatility to the conduct of monetary policy. Building on a stylized model, I show empirically that monetary policy may have been `too successful' in stabilizing inflation, as this has contributed to excessive financial risk taking.