Reliable methods for analysing and managing the sources of dependence between financial assets at all frequencies (FWOTM815)

Is a derivative correctly priced? Is the financial market sufficient liquid such that individual trades have no substantial price impact? How can a portfolio be constructed in order to diversify away the extreme risks of large negative returns and is it possible to include jump risk in its Value-at-Risk estimation? These are all important financial questions for which fundamental research is needed to disentangle the normal from the non-normal and the common from the idiosyncratic risk factors, driving financial asset returns. The past years have shown that the methodologies used to assess portfolio risk in the industry lead to severe underestimation of potential losses. The goal of this project is to gain a better understanding of the multivariate dependence structure connecting the financial returns of the multitude of financial assets traded in financial markets, observed at both the ultra-high (intraday) frequency and the lower frequency of weekly and monthly observations. Distinguishing the different risk factors is key to obtaining answers to questions such as the ones posed above. Practitioners will benefit from more accurate valuation of financial assets and better risk management tools. Besides, the estimators working on high frequency data will allow to research the impact of liquidity on price jumps in the market. Our proposed estimators will be tested and compared with available methods in extensive simulation studies and on real-life financial data.

Keywords:accounting, Financial Analysis

Disciplines:Business administration and accounting not elsewhere classified