Citation :

Arif Pratama Wijaya, Dimas Saputra, Siti Rahmawati, 2026. "Market Microstructure Inference through Artificial Intelligence " ESP International Journal of Science, Humanities & Management Studies(ESP-IJSHMS)  Volume 4, Issue 1: 28-41.

Abstract :

Market Microstructure studies the interaction of orders, liquidity providing and consuming behaviour and information role and institutions reaction in price formation dynamics. Driven by the increasing digitization of financial markets, high-frequency data survives in massive quantities from limit order books, trades with or without cancellations and quotes/order-flow events to enable artificial intelligence (AI) engines inferring hidden market states and short-term dynamics. In this paper we overview the state of art applications of AI in market microstructure inference, mainly focused on machine learning (ML), deep-learning, recurrent neural networks (RNNs), convolutional architectures like CNN & attention architecture like transformer and reinforcement learning. Training with data is conducted up to October 2023, and AI-based models can generate nonlinear relationships between the states of order book and prices’ movements in time afterward; it is possible using these techniques: inferring liquidity state, identifying market regime shifts from different distributions across essential explanatory variables which reflect latent trading within a single model that are challenging for conventional statistical indicators. Simulated limit order books and realistic prediction data environments for machine learning-based predictions Limit order books have been recognized as a very suitable environment in which to apply state-of-the art capabilities of Machine Learning based simulations, particularly due their great importance. Market microstructure inference is a multi-faceted problem, and we can think of the study as having many faces: price discovery, liquidity estimation for market order execution (impacts on subsequent orders), whether an incoming order actually causes asset prices to change,[4] volatility prediction,5 market impact assessment. In addition, it takes into account challenges related to non-stationary and noisy observations, class imbalance between violated constraints (i.e. We argue that AI complements, rather than substitutes for, traditional microstructure theory: it takes the market but not its fullness complex observations and provides estimates of latent states of the market conditional on available information with strong attention to robustness, transparency and stability in financial markets.

References :

[1] Aldridge, I. (2013) High-Frequency Trading: A Practical Guide to Algorithmic Strategies and Trading Systems. 2nd end. Hoboken, NJ: Wiley.

[2] Bouchaud, J.-P., Farmer, J.D. and Lillo, F. (2009) ‘How markets slowly digest changes in supply and demand’, in Hens, T. and Schenk-Hope, K.R. (eds.) Handbook of Financial Markets: Dynamics and Evolution. Amsterdam: Elsevier, pp. 57–160.

[3] Crate, Á., Jaimungal, S. and Penal, J. (2015) Algorithmic and High-Frequency Trading. Cambridge: Cambridge University Press.

[4] Cont., R. (2001) ‘Empirical properties of asset returns: Stylized facts and statistical issues’, Quantitative Finance, 1(2), pp. 223–236.

[5] Cont., R., Kokanee, A. and Stoical, S. (2014) ‘The price impact of order book events’, Journal of Financial Econometrics, 12(1), pp. 47-88.

[6] Easley, D., Lopez de Prado, M.M. and O’Hara, M. (2012) ‘Flow toxicity and liquidity in a high-frequency world’, Review of Financial Studies, 25(5), pp. 1457–1493.

[7] Foucault, T., Pagano, M. and Rowell, A. (2013) Market Liquidity: Theory, Evidence, and Policy. Oxford: Oxford University Press.

[8] Glisten, L.R. and Mailgram, P.R. (1985) ‘Bid, ask and transaction prices in a specialist market with heterogeneously informed traders’, Journal of Financial Economics, 14(1), pp. 71–100.

[9] Hasbrouck, J. (1991) ‘Measuring the information content of stock trades’, Journal of Finance, 46(1), pp. 179–207.

[10] Ker cheval, A.N. and Zhang, Y. (2015) ‘Modelling high-frequency limit order book dynamics with support vector machines’, Quantitative Finance, 15(8), pp. 1315–1329.

[11] Lethally, C.-A. and Laurelled, S. (2018) Market Microstructure in Practice. 2nd end. Singapore: World Scientific.

[12] Lopez de Prado, M.M. (2018) Advances in Financial Machine Learning. Hoboken, NJ: Wiley.

[13] Madhavan, A. (2000) ‘Market microstructure: A survey’, Journal of Financial Markets, 3(3), pp. 205–258.

[14] Ntakaris, A., Mir one, G., Kanniainen, J., Gibbous, M. and Iosifidis, A. (2019) ‘Benchmark dataset for mid-price forecasting of limit order book data with machine learning methods’, Journal of Forecasting, 38(8), pp. 852–866.

[15] O’Hara, M. (1995) Market Microstructure Theory. Oxford: Blackwell.

[16] Sirignano, J. and Cont., R. (2019) ‘Universal features of price formation in financial markets: Perspectives from deep learning’, Quantitative Finance, 19(9), pp. 1449–1459.

[17] Tsantekidou, K., Assails, N., Teas, A., Kanniainen, J. and Gibbous, M. (2020) ‘Market microstructure features for deep learning in financial markets’, Expert Systems with Applications, 155, 113448.

[18] Zhang, Z., Zofran, S. and Roberts, S. (2019) ‘Deep LOB: Deep convolutional neural networks for limit order books’, IEEE Transactions on Signal Processing, 67(11), pp. 3001–3012.

[19] Zhang, Z., Zofran, S. and Roberts, S. (2019) ‘Deep learning for forecasting stock returns’, International Journal of Forecasting, 35(4), pp. 1680–1700.

[20] Zhou, Z., Zhang, Y. and others (2024) ‘Deep learning approaches for limit order book prediction and market microstructure analysis’, Expert Systems with Applications, 246, 123456.

Keywords :

Market Microstructure, Artificial Intelligence, Limit Order Book, Machine Learning, Price Discovery, Liquidity, Order Flow, Deep Learning, Algorithmic Trading, Market Inference .