Algorithmic trading | EPFL Graph Search

Algorithmic trading is a method of executing orders using automated pre-programmed trading instructions accounting for variables such as time, price, and volume. This type of trading attempts to leverage the speed and computational resources of computers relative to human traders. In the twenty-first century, algorithmic trading has been gaining traction with both retail and institutional traders. A study in 2019 showed that around 92% of trading in the Forex market was performed by trading algorithms rather than humans. It is widely used by investment banks, pension funds, mutual funds, and hedge funds that may need to spread out the execution of a larger order or perform trades too fast for human traders to react to. However, it is also available to private traders using simple retail tools. The term algorithmic trading is often used synonymously with automated trading system. These encompass a variety of trading strategies, some of which are based on formulas and results from ma

Three Essays on Corporate Disclosure

Evgeny Petrov

Corporate disclosure is the most important source of information about the firm for the outside investors. While some disclosure of public firms is mandated by regulation, firm managers can provide extra information at their discretion by making voluntary disclosures. On the other hand, even the reports required by regulation can be disclosed untruthfully. This thesis is structured in three chapters, each addressing a specific issue in voluntary disclosure and misreporting. In the first chapter, titled

Voluntary Disclosure and Informed Trading'', I study the impact of informed trading on voluntary corporate disclosure in the presence of two frictions: cost of disclosure and value of manager's information. In the absence of both frictions, informed trading has no impact on disclosure even when traders are not certain whether the manager has information. If disclosure is costly, then informed trading reduces disclosure. Since traders can discover favorable information about the firm, additional disclosure of the information is not necessary. If manager's information is valuable for the firm, then informed trading increases disclosure. Since traders can discover unfavorable information about the firm, the manager with such information has less incentives to pool with uninformed managers and discloses to show that he is informed. I also show that informed trading can have both a positive and a negative real effect on the firm value by crowding in or crowding out information production in the firm. These results hold for general information structures and are robust if traders can choose how much information to acquire. The second chapter, titled

Misreporting and Feedback Effect'' and co-authored with Prof. Hui Chen of the University of Zurich, studies the incentives of firms to misreport information in the presence of feedback effect from financial markets. Stock price often provides firms with new information, which can be used in the firms' subsequent real decisions. We examine how this informational feedback from the financial market affects a myopic firm manager's incentive to misreport, and how the misreporting further affects the firm's price and value. We find that the manager overstates his report more in the presence of feedback, but this misreporting brings forth both positive price and real effects for the firm. Intuitively, overstating the report encourages information production in the market because (a) it renders accounting reports less reliable as a source of information, and (b) investors expect higher trading profits from larger capital investment. The new incremental information improves investment efficiency when it is revealed to the firm manager through trading and used in the firm's subsequent investment decisions. As a consequence, the capital investment is higher when there is feedback effect. In the third chapter, titled ``Voluntary Disclosure and Margin Constraints'', I develop a dynamic model of voluntary corporate disclosure that explains clustering of negative announcements observed in practice. A manager may receive a signal about the firm's asset value and can disclose it to traders with margin constraints. I show that the manager postpones delivery of a negative signal until the margin constraints tighten. In contrast to previous studies, the clustering of announcements happens even if there are no negative updates in traders' beliefs about the firm value.

EPFL2017

Trading mechanisms in over-the-counter markets

Sebastian Vogel

In the chapter

When to Introduce Electronic Trading Platforms in Over-the-Counter Markets?'' An equilibrium in a market is determined in which traders have the choice between using an electronic platform with a request-for-quote protocol or calling a dealer directly. The main takeaway is that platforms increase the bargaining power of traders with search costs, thereby increasing their market participation and potentially also welfare. Another friction that affects trading in OTC markets are informational frictions. market participants may learn from trading activity on platforms and trade on that information in other trading venues. The interplay between an electronic trading platform in the dealer-to-customer market and the interdealer market is analysed in the chapter

Electronic Trading in OTC Markets vs. Centralized Exchange.'' In that chapter, a two-tiered OTC market is compared with a fully centralized market, showing that informed traders may prefer the two-tiered structure in order to benefit from their information. This ability of decentralized OTC markets to deal with asymmetric information is also a major benefit of OTC markets compared to exchange markets in welfare considerations. In the chapter ``Informed Traders and Dealers in the FX Forward Market,'' it is analyzed whether such informational frictions exist, whether dealers are able to mitigate those by charging different markups to different clients and whether we behavior consistent with the alleviation of informational frictions.

EPFL2020

Essays in Empirical Asset Pricing

Alexis Arilès Marchal

This thesis consists of three applications of machine learning techniques to empirical asset pricing.In the first part, which is co-authored work with Oksana Bashchenko, we develop a new method that detects jumps nonparametrically in financial time series and significantly outperforms the current benchmark on simulated data. We use a long short-term memory (LSTM) neural network that is trained on labelled data generated by a process that experiences both jumps and volatility bursts. As a result, the network learns how to disentangle the two. Then it is applied to out-of-sample simulated data and delivers results that considerably differ from the benchmark: we obtain fewer spurious detection and identify a larger number of true jumps. When applied to real data, our approach for jump screening allows to extract a more precise signal about future volatility.In the second part, which is co-authored work with Oksana Bashchenko, we develop a methodology for detecting asset bubbles using a neural network. We rely on the theory of local martingales in continuous-time and use a deep network to estimate the diffusion coefficient of the price process more accurately than the current estimator, obtaining an improved detection of bubbles. We show the outperformance of our algorithm over the existing statistical method in a laboratory created with simulated data. We then apply the network classification to real data and build a zero net exposure trading strategy that exploits the risky arbitrage emanating from the presence of bubbles in the US equity market from 2006 to 2008. The profitability of the strategy provides an estimation of the economical magnitude of bubbles as well as support for the theoretical assumptions relied on.In the third part, I propose a new tool to characterize the resolution of uncertainty around FOMC press conferences. It relies on the construction of a measure capturing the level of discussion complexity between the Fed Chair and reporters during the Q&A sessions. I show that complex discussions are associated with higher equity returns and a drop in realized volatility. The method creates an attention score by quantifying how much the Chair needs to rely on reading internal documents to be able to answer a question. This is accomplished by building a novel dataset of video images of the press conferences and leveraging recent deep learning algorithms from computer vision. This alternative data provides new information on nonverbal communication that cannot be extracted from the widely analyzed FOMC transcripts. This chapter can be seen as a proof of concept that certain videos contain valuable information for the study of financial markets.

EPFL2022