An empirical essay in Forex market

Li, Dangyang (2021) An empirical essay in Forex market. PhD thesis, University of Glasgow.

Abstract

The Forex market is the largest financial market globally, with daily trading volumes of about 6.6 trillion U.S. dollars per day in April 2019 (Bank for International Settlements, (BIS) 2019). The forex market always attracts the view of many researchers. This doctoral thesis particularly pays attention to some popular forex market questions.

The first chapter provides a comprehensive review of recent developments of forex market researches. The first section of the chapter briefly introduces the currenciesspecific pricing factors and the forex market portfolio management. The second part gives an introduction to the risk forecasting model in the financial market. Finally, I will illustrate the concept of the forward premium puzzle and interest rate parity in the last part and review the literature on the forward premium puzzle.

In the second chapter, I focus on the prevalent forex factors’ tail dependence. Lustiget al. (2011) have introduced some forex factors, the ’dollar risk factor’ (DOL) and the ’carry trade factor’ (HML), and shown that they can price carry trade portfolios in the cross-section. This new result is helpful in the academic literature on cross-sectional asset pricing, risk management, and portfolio optimization. Those factors are also widely used in the industry. This thesis tests the relevance of four popular currency specific factors contributing to a diversified forex portfolio and risk management. I show that modeling non-linear dependency across the factors is essential and adds value to a forex portfolio.

Then, I concentrate on the risk forecasting of the forex factor portfolios, which I discussed in the previous chapter. Risk forecasting is a popular research question in the financial market. Value at Risk (VaR) is the standard measure of risk, which is the common risk measure when forecasting the risk. Meanwhile, the expected shortfall (ES) is the conditional expectation of exceeding beyond the VaR. I will apply the Copula to improve the risk forecast model of Patton et al. (2019) to forecast the VaR and ES of factors portfolio returns. I apply the goodness-of-fit test and the Diebold-Mariano tests to evaluate the performance of the forecasting models. The results show that the Copula multivariate dynamic forecasting models have their benefit when estimating the future risk.

In the last chapter, I present a model, following Burnside et al. (2009), which apply the adverse selection problem between a market maker and trader rationalized, to discuss the negative covariance between the forward premium and spot rate change. I first apply the unique order flow data set to estimate Burnside et al. (2009) model. Then, I creatively discuss the bond and spot exchange transactions to explain the excess return of carry trade. According to the estimation results, I could conclude that the adverse selection could generate the forward premium puzzle.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Colleges/Schools: College of Social Sciences > Adam Smith Business School
• Supervisor's Name: Cerrato, Professor Mario and Burnside, Professor Craig
• Date of Award: 2021
• Depositing User: Theses Team
• Unique ID: glathesis:2021-82595
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 17 Dec 2021 09:51
• Last Modified: 08 Apr 2022 17:02
• Thesis DOI: 10.5525/gla.thesis.82595
• URI: https://theses.gla.ac.uk/id/eprint/82595