invert-default-slider-image

Enrique Martinez Miranda

Home »  Enrique Martinez Miranda

 

EnriquePhotoProfile

Enrique has Masters degree in Engineering and bacherlors in Physics, both from the National Autonomous University of Mexico (UNAM). Currently, he is a first year postgraduate research student at the Department of Informatics in King's College London, under supervision of Dr. Matthew Howard and Professor Peter McBurney

 

Research Project

Currently, his research is based on the description of market manipulators based on the trading process, specifically in spoofing and pinging strategies and how these can be modeled within a reinforcement learning framework. In the future, this research project will evolve to make a broader analysis of market manipulation, from a behavioral description up to the development of new market manipulation detector.

Project Update

pricechangeAt this moment, Enrique has been interested in the prediction of the process of price manipulation by spoofing in the financial markets. Basically, spoofing intends to manipulate the price of a given asset by breaking the supply and demand equilibrium, so other traders playing in the market will be influenced by this wrong information and the market, as a whole, will react by changing the prices.

lob

Enrique has developed a deterministic model for the dynamics of the so-called Limit Order Book (LOB).The LOB is a system used in modern financial markets where buyers and sellers place their orders. It basically represents the supply and demand and the price changes according to how the orders are being placed, executed or cancelled.

realvspredictionThe market price is merely determined by the average price between the best bid and offer quotes. So far, the model tries to replicate the dynamics of the LOB after a trader submits, for example, a large sell order. Once this happens, given the supply and demand law it is expected to see a drop in the prices as the buy side may believe a large trader has, for example, information about the asset. The LOB model is based on a Finite Impulse Response (FIR) model and given the state of the LOB, the model tries to replicate the trend in the volumes at each of the price levels of the LOB, just like it is shown in the Figure at the right side.

Research Interest

  • (Inverse) Reinforcement learning
  • Market manipulation
  • Algorithmic trading
  • High-frequency trading

Publications

  • Learning unfair trading: a market manipulation analysis from the reinforcement learning perspective. E. Martinez-Miranda, P. McBurney, and M. Howard. IEEE International Conference on Evolving and Adaptive Intelligent Systems, 2016. In Press.
  • Gordillo-Ruiz, J.L., Martínez-Miranda, E., and Stephens, C.R., "Develando Estrategias de Mercado: Minería de Datos Aplicada al Análisis de Mercados Financieros", Computación y Sistemas 16 (2012), no. 2, 221-231.
  • Stephens, C.R., Gordillo, J.L., Martínez-Miranda, E.: "Who's Smart and Who's Lucky? Inferring Trading Strategy, Learning and Adaptation in Financial Markets through Data Mining", Brabazon, Anthony and O'Neill, Michael (eds.): Natural Computing in Computational Finance (Springer-Verlag, 2009).

Contact

Department of Informatics, 6th Floor, Desk D27

Strand Campus

King's College London, Strand,

London, WC2R 2LS

United Kingdom

Email: enrique[dot]martinez_miranda@kcl.ac.uk