Imperial College London


Business School

Senior Research Fellow



+44 (0)20 7594 9089a.kirilenko




4.0852-53 Prince's GateSouth Kensington Campus





Publication Type

12 results found

Adamic L, Brunetti C, Harris JH, Kirilenko Aet al., 2017, Trading networks, The Econometrics Journal, Vol: 20, Pages: S126-S149, ISSN: 1368-4221


Burn L, Faull J, Kirilenko A, Laffan B, Landau J, Manuelides Y, Micheler E, Miles D, Sandbu M, Smolenska A, Schlosser P, Zettelmeyer Jet al., 2017, The changing geography of finance and regulation in Europe, Publisher: European University Institute, ISBN: 9789290845454


KIRILENKO A, KYLE AS, SAMADI M, TUZUN Tet al., 2017, The Flash Crash: High-Frequency Trading in an Electronic Market, The Journal of Finance, Vol: 72, Pages: 967-998, ISSN: 0022-1082


Yang SY, Mo SYK, Liu A, Kirilenko AAet al., 2017, Genetic programming optimization for a sentiment feedback strength based trading strategy, Neurocomputing, Vol: 264, Pages: 29-41, ISSN: 0925-2312

© 2017 Elsevier B.V. This study is motivated by the empirical findings that news and social media Twitter messages (tweets) exhibit persistent predictive power on financial market movement. Based on the evidence that tweets are faster than news in revealing new market information, whereas news is regarded broadly a more reliable source of information than tweets, we propose a superior trading strategy based on the sentiment feedback strength between the news and tweets using generic programming optimization method. The key intuition behind this feedback strength based approach is that the joint momentum of the two sentiment series leads to significant market signals, which can be exploited to generate superior trading profits. With the trade-off between information speed and its reliability, this study aims to develop an optimal trading strategy using investors’ sentiment feedback strength with the objective to maximize risk adjusted return measured by the Sterling ratio. We find that the sentiment feedback based strategies yield superior market returns with low maximum drawdown over the period from 2012 to 2015. In comparison, the strategies based on the sentiment feedback indicator generate over 14.7% Sterling ratio compared with 10.4% and 13.6% from the technical indicator-based strategies and the basic buy-and-hold strategy respectively. After considering transaction costs, the sentiment indicator based strategy outperforms the technical indicator based strategy consistently. Backtesting shows that the advantage is statistically significant. The result suggests that the sentiment feedback indicator provides support in controlling loss with lower maximum drawdown.


Cheng I-H, Kirilenko A, Xiong W, 2015, Convective Risk Flows in Commodity Futures Markets*, Publisher: OXFORD UNIV PRESS, Pages: 1733-1781, ISSN: 1572-3097


Yang SY, Qiao Q, Beling PA, Scherer WT, Kirilenko AAet al., 2015, Gaussian process-based algorithmic trading strategy identification, Quantitative Finance, Vol: 15, Pages: 1683-1703, ISSN: 1469-7688

© 2015 Taylor & Francis. Many market participants now employ algorithmic trading, commonly defined as the use of computer algorithms, to automatically make certain trading decisions, submit orders and manage those orders after submission. Identifying and understanding the impact of algorithmic trading on financial markets has become a critical issue for market operators and regulators. Advanced data feeds and audit trail information from market operators now allow for the full observation of market participants’ actions. A key question is the extent to which it is possible to understand and characterize the behaviour of individual participants from observations of trading actions. In this paper, we consider the basic problems of categorizing and recognizing traders (or, equivalently, trading algorithms) on the basis of observed limit orders. These problems are of interest to regulators engaged in strategy identification for the purposes of fraud detection and policy development. Methods have been suggested in the literature for describing trader behaviour using classification rules defined over a feature space consisting of summary trading statistics of volume and inventory, along with derived variables that reflect the consistency of buying or selling behaviour. Our principal contribution is to suggest an entirely different feature space that is constructed by inferring key parameters of a sequential optimization model that we take as a surrogate for the decision-making process of the traders. In particular, we model trader behaviour in terms of a Markov decision process. We infer the reward (or objective) function for this process from observations of trading actions using a process from machine learning known as inverse reinforcement learning (IRL). The reward functions learned through IRL then constitute a feature space that can be the basis for supervised learning (for classification or recognition of traders) or unsupervised learning (for cate


Cohen-Cole E, Kirilenko A, Patacchini E, 2014, Trading networks and liquidity provision, JOURNAL OF FINANCIAL ECONOMICS, Vol: 113, Pages: 235-251, ISSN: 0304-405X


Elliott M, Golub B, Kirilenko A, 2014, How Sharing Information Can Garble Experts' Advice, 126th Annual Meeting American-Economic-Association, Publisher: AMER ECONOMIC ASSOC, Pages: 463-468, ISSN: 0002-8282


Kirilenko A, Mankad S, Michailidis G, 2014, Do U.S. regulators listen to the public? Testing the regulatory process with the RegRank algorithm, ISSN: 0730-8078

We propose a tool called RegRank that can be used to measure and test whether government regulatory agencies adjust aspects of final rules in response to comments received from the public. The algorithm, which combines customized dictionaries with LDA topic models, is used to analyze the text of public rulemaking documents of the Commodity Futures Trading Commission (CFTC) - a federal regulatory agency in charge of implementing parts of the Dodd-Frank Wall Street Reform and Consumer Protection Act. A key finding based on the available data is that the government adjusts its final rules in the direction of public comments.


Kirilenko A, Sowers RB, Meng X, 2013, A multiscale model of high-frequency trading, Algorithmic Finance, Vol: 2, Pages: 59-98, ISSN: 2158-5571

© 2013 - IOS Press and the authors. We propose and study a stylization of high frequency trading (HFT). Our interest is an order book which consists of orders from slow liquidity traders and orders from high-frequency traders. We would like to frame a model which is amenable to the (seemingly natural) mathematical toolkit of separation of scales and which can be used to address some of the larger issues involved in HFT. The main issue to which we address our model is volatility. An important question is how volatility is affected by HFT. In our stylized model, we show how HFT increases volatility, and can quantify this effect as a function of the parameters in our model and the separation of scales.


Kirilenko AA, Lo AW, 2013, Moore's law versus Murphy's law: Algorithmic trading and its discontents, Pages: 51-72


Mankad S, Michailidis G, Kirilenko A, 2013, Discovering the ecosystem of an electronic financial market with a dynamic machine-learning method, Algorithmic Finance, Vol: 2, Pages: 151-165, ISSN: 2158-5571

© 2013 - IOS Press and the authors. All rights reserved. Not long ago securities were traded by human traders in face-to-face markets. The ecosystem of an open outcry market was well-known, visible to a human eye, and rigidly prescribed. Now trading is increasingly done in anonymous electronic markets where traders do not have designated functions or mandatory roles. In fact, the traders themselves have been replaced by algorithms (machines) operating with little or no human oversight. While the process of electronic trading is not visible to a human eye, machine-learning methods have been developed to recognize persistent patterns in the data. In this study, we develop a dynamic machine-learning method that designates traders in an anonymous electronic market into five persistent categories: high frequency traders, market makers, opportunistic traders, fundamental traders, and small traders. Our method extends a plaid clustering technique with a smoothing framework that filters out transient patterns. The method is fast, robust, and suitable for a discovering trading ecosystems in a large number of electronic markets.


This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: Request URI: /respub/WEB-INF/jsp/search-html.jsp Query String: respub-action=search.html&id=00883159&limit=30&person=true