This thesis contains three empirical studies on the US corporate bond market; each chapter is self-contained and can be read independently. Chapter 1 studies the impact of credit rating changes on corporate bond returns. This study uses a large dataset of corporate bond transactions from the TRACE database for the US corporate bond market, combined with credit rating changes from Fitch, Moody's and Standard and Poor's (S&P), to analyse over 22,000 bonds, coupled with approximately 28,400 rating events over nearly six years. The results show that the bond market responds to news on credit quality asymmetrically: credit rating downgrades, representing bad news for bond holders, produce the strongest response in returns, whilst upgrades do not generate a statistically significant increase in returns. Chapter 2 analyses how order flow (investor "buy" and "sell" trades), impacts corporate bond prices. Order flow plays an important informational role, acting as a conduit through which private information about fundamental value is aggregated into prices. Using intraday transaction data from the TRACE database, I analyse over 1,000 of the most liquid corporate bonds, a total of 9.5 million trades. Drawing on similar studies of other markets, the relationship between returns and order flow is modelled using a vector autoregression, and the information content of a trade is measured as the long-run price impact of a shock to order flow. Price impacts are particularly strong and significant for order flow from institutional investors and for bonds with higher default risk, higher volatility and lower liquidity. Chapter 3 provides novel evidence on the importance of high frequency measures of volatility and correlation for the corporate bond market. Realized measures of volatility have been shown to be important in modelling and forecasting equity, exchange rate, and Treasury bill return volatility. We merge the NYSE's TAQ database of high frequency equity prices with the TRACE database, and show that the information contained in high frequency data is valuable in modelling the dynamics of the firm-level covariance matrix of bond and stock returns, for over 100 individual U.S. firms.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:729133 |
Date | January 2017 |
Creators | Levonmaa, Aino |
Contributors | Sheppard, Kevin |
Publisher | University of Oxford |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | https://ora.ox.ac.uk/objects/uuid:488f7a83-749f-4466-9bc5-e7a558edb6bf |
Page generated in 0.0015 seconds