MarkMH.github.io/hft_thesis.html at main · MarkMH/MarkMH.github.io · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
<!DOCTYPE HTML>
<!--
	Basic structure of Strata from HTML5 UP html5up.net | @ajlkn
	Free for private and commercial use under CCA 3.0 license (html5up.net/license)
	Significant customizations by Mark Marner-Hausen
-->
<html>
	<head>
		<title>Webpage Marner-Hausen</title>
		<meta charset="utf-8" />
		<meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=no" />
		<link rel="stylesheet" href="assets/css/main.css" />
	</head>
	<body class="is-preload">

		<!-- Header -->
			<header id="header">
				<div class="inner">
					<a class="image avatar"><img src="images/IMG_9947.jpg" alt="" /></a>
					<h1><strong>Hi 👋, I'm Mark,<br />
						a Senior ML Engineer <br />
						with scientific background</strong></a></h1>
				</div>
			</header>

		<!-- Main -->
			<div id="main">

				<!-- One -->
					<section id="one">
						<header class="major">
							<h1>Developing a Framework for Real-Time Trading in a Laboratory Financial Market</h1>

							<p align="justify" style="text-align:justify"><b>
								This project deals with high-frequency trading and economic engineering means to prevent arbitrage
								profits in the financial market. These quasi-risk-free profits are only accessible to so-called high-frequency
								traders (HFTs), such as BlackRock and Vanguard. HFTs are market participants typically characterized by
								algorithmic trading at the speed of light without any intrinsic interest to own or owe an asset.
								The basic idea is that simultaneously available
								information that affects stock prices - think of a tweet from Elon Musk or Donald Trump - might be exploited by
								those who react first. This raises concerns in both academia and public policy.
							</b></p>

							<p align="justify" style="text-align:justify"><b>
								Although Economic engineering offers approaches to eliminate these arbitrage opportunities,
								regulators face the difficult task of identifying appropriate market interventions without having
								landmark data. Data is impossible to collect since it is more than unlikely that financial markets around the
								world will change their rules for a day just for the sake of experimentation. But indeed, this
								would be necessary, as the solution approaches only realize their full potential if there is a world-wide rollout.
								Therefore, we are testing several of the market design solution in an artificial financial market in a
								laboratory. This artificial market enables real-time algorithmic trading at millisecond speeds, providing a
								more accurate laboratory replication of the financial market mechanisms relevant to high-frequency trading
								than has been achieved up to this point. However, this project discussed the difficulties of developing
								difficult a sufficiently sophisticated software, and moreover, it is far from trivial to adequately communicate
								the complex financial market rules to our non-expert lab subjects. To overcome these challenges, this project
								discusses bottlenecks arising from the underlying software and the laboratory infrastructure, plus it analyzes
								data from pilot sessions.
							</b></p>

							<p align="justify" style="text-align:justify"><b><span class="image left"><img src="images/fulls/ui_hft.png" alt="" /></span>
								The image on the left shows the trading platform available to subjects during each trading period. Subjects
								a given the choice to take on the role of HFTS by turning on a trading bot, which will trade on their behalf.
								The underlying artificial exchange ensures serial processing of messages, an important part of the real-world financial
								market, but a difficult function to implement in the lab given modern multi-core processors.
							</b></p>

							<p align="justify" style="text-align:justify"><b><span class="image right"><img src="images/fulls/hft_prices.png"  /></span>
								Another major advantage of the laboratory study, apart from world-wide roll-out, is the full control over price
								movements. This allows us to study the behavior of HFTs under both normal and stressful market conditions. Stressful
								market conditions can take different forms, for example, the left side of this figure shows a sudden and significant
								market collapse, while the right side shows a very volatile market. In fact, even a boom phase, such as the recent
								GameStop (GME) frenzy in early 2021, may be considered stressful for HFTs.
							</b></p>

							<p align="justify" style="text-align:justify"><b>
								If you are interested in more particulars and technical details, please feel free to check out my Working Paper now
								available online. If you are more interested in the data analysis of the pilots and the lessons learned,
								please checkout GitHub.
							</b></p>
						</header>
					</section>

					<section>
								<ul class="actions fit">
									<li><a href="https://www.econtribute.de/RePEc/ajk/ajkdps/ECONtribute_172_2022.pdf" class="button fit">Working Paper</a></li>
									<li><a href="https://github.com/MarkMH/hft_data_analysis" class="button fit"><i class="icon brands fa-github"></i> Browse Code
									</a></li>
									<li><a href="index.html" class="button fit">Portfolio</a></li>
								</ul>
					</section>
			</div>

		<!-- Footer -->
			<footer id="footer">
				<div class="inner">
					<ul class="icons">
						<br>
						<li><a href="https://github.com/MarkMH" class="icon brands fa-github"><span class="label">Github</span></a></li>
						<li><a href="https://www.linkedin.com/in/mark-marner-hausen-854a3a117/" class="icon brands fa-linkedin"><span class="label">LinkedIn</span></a></li>
					</ul>
					<ul
					<li><br> Mark Marner-Hausen <br />
						</p>
					</ul>
				</div>
			</footer>

		<!-- Scripts -->
			<script src="assets/js/jquery.min.js"></script>
			<script src="assets/js/jquery.poptrox.min.js"></script>
			<script src="assets/js/browser.min.js"></script>
			<script src="assets/js/breakpoints.min.js"></script>
			<script src="assets/js/util.js"></script>
			<script src="assets/js/main.js"></script>

	</body>
</html>