Advanced Modelling Topics 2

The next course will take place: Tuesday 19 - Friday 22 February 2019

 

Syllabus

  • Algorithmic and high frequency trading
  • Limit order books and market microstructure: The field of market microstructure is concerned with the study of financial markets on the microscopic scale. Thanks to the availability of high-frequency data that describes the temporal evolution of financial markets at the level of individual order arrivals and departures, the study of market microstructure has recently provided many new insights into several long-standing questions on diverse topics such as market efficiency, market stability, and the sources of volatility. The field is also highly relevant from a practical perspective, because a detailed understanding of market microstructure helps practitioners to design efficient execution strategies and to improve their estimation of risk exposure. In this course, we will study how several widely observed but highly non-trivial mesoscopic- and macroscopic-scale properties of financial markets emerge from the microscopic-scale actions and interactions of individual traders. We will study in detail the process of trading via a limit order book, and contrast this mechanism to both open-outcry and quote-driven trading. We will introduce a mathematical framework for studying the temporal evolution of a limit order book, use this framework to discuss two recent limit order book models, and discuss how such models can help to illuminate the delicate interplay between order flow, liquidity, and price formation. Finally, we will observe that many properties of financial markets that were previously regarded as a direct result of traders' strategic actions may in fact emerge as a natural consequence of market microstructure.
  • Energy markets: We begin by reviewing the workings of energy markets, highlighting features that distinguish them from other financial markets. We then look at modelling approaches:  basic models that incorporate features such as mean-reversion, seasonality and jumps, and various kinds of multi-factor models that aim to capture forward price dynamics. Finally we consider a range of energy contracts and assets and discuss valuation and risk management techniques.
  • Machine Learning Fundamentals
  • Robust Methods: The standard approach to option pricing is to propose a probabilistic model for the dynamics of the underlying price process and to use this model to derive prices for vanilla options (e.g. call and put) or exotic options. The starting point of this lecture is to drop the assumption of a probabilistic model fully describing the price evolution. Instead, we take only prices of (some) liquidly traded options as exogenously given by the market, and use those prices to learn about the properties of the underlying asset prices and of non-liquidly traded options (typically exotic options). As we will see, knowing the prices of traded derivatives gives constraints on the law of the asset price and such on the prices of non-liquidly traded derivatives. Our aim is then to find the best possible model-independent bounds on the arbitrage-free prices for various exotic options, taking the prices of liquidly traded derivatives as given.

Please note that in exceptional circumstances it may be necessary to cancel or alter a particular lecture, so that these details are subject to small variation.

 

For students enrolled on course

Course Materials - including student instructions, lecture notes, assignment and submission link