Machine Learning Models Can Help Reduce Volatility and Boost Stock Market Returns
The use of machine learning models makes it possible to achieve greater accuracy in predicting risks in the Russian stock market compared to classical econometric approaches. The predictive power of these models increases by 23%, while the average investor’s return can reach up to 13% per annum. These conclusions were drawn by Nikita Lysenok from the Department of Financial Market Infrastructure at the HSE Faculty of Economic Sciences. The paper has been published in Fundamental and Applied Mathematics.
Volatility assessment remains a central task in investment risk management and stock market strategy development. It measures the degree of asset price fluctuation, indicating how much a price can vary during a trading day. Errors in estimating future volatility directly affect option pricing, portfolio risk management, and the overall effectiveness of investment strategies.
There are traditional methods for estimating volatility, such as the so-called HAR model. It evaluates three time horizons based on specified parameters and produces a linear forecast of next-day volatility. However, this model does not perform well when assessing nonlinear risks. The author therefore set out to determine whether machine learning methods could deliver better results than the traditional econometric approach.
The study is based on high-frequency data for the ten most liquid stocks traded on the Moscow Stock Exchange from 2014 to 2025. Price changes at 10-minute intervals were analysed, making it possible to calculate realised volatility and its variations for subsequent forecasting. In addition, variables reflecting return characteristics and market structure were incorporated to capture the specific features of trading on the Russian stock exchange. The results of the classical HAR model were then compared with those of machine learning methods, including Random Forest, XGBoost, and LightGBM, which can account for non-linearities and complex interactions among variables. To evaluate the forecast quality, the author not only applied standard statistical metrics but also tested the predicted volatility in a simulation of actual stock trading.
Nikita Lysenok
The results demonstrated a consistent advantage of the machine learning models over the classical HAR model. The most significant improvements were observed in short- and medium-term forecasts. Average forecast errors decreased by approximately 15%, while prediction accuracy increased by up to 23%, depending on the method, with LightGBM delivering the best performance. 'Machine learning models demonstrate superior performance during both stable market conditions and periods of heightened volatility. While they do not possess the gift of foresight, they can more accurately signal rising risks, providing investors with time to adjust and rebalance their portfolios,' explains Nikita Lysenok from the Department of Financial Market Infrastructure at the HSE FES.
A simulation of stock trading using enhanced volatility estimates from machine learning showed an increase in annual profitability of 7 percentage points, rising from 6.48% to 13.68%. 'This is where the cumulative effect and risk asymmetry come into play. More accurate volatility estimates allow for optimal position sizing. The difference may be small for a single trade, but over hundreds of trades, it translates into substantial gains in annual returns,' explains Lysenok.
At the same time, the author emphasised that machine learning methods for predicting volatility require careful parameter tuning and are highly dependent on the quality of the training data. Higher accuracy in forecasting volatility does not automatically translate into economic gains, and trading still demands considerable skill and involves significant risk. 'A key factor in sustainable profitability is risk management: accurate volatility forecasts allow for timely position adjustments and help avoid severe drawdowns,' the author notes.
See also:
HSE Computer Science Researchers Win Gold Medal at International Machine Learning Competition
A team comprising HSE International Laboratory of Statistical and Computational Genomics researchers Aleksei Shmelev and Nikita Chervov, 2025 graduate of the HSE Faculty of Computer Science’s Master’s programme in Data Analysis in Biology and Medicine Ivan Gevorkov, and two students from the United States achieved an outstanding result at the 2026 NeuroGolf international machine learning championship. The team won a gold medal and placed seventh overall.
‘Working with AI Solves a Wide Range of Engineering Problems’
Artificial intelligence is a working tool based on a balanced combination of algorithms and engineering. Experts and doctoral students from the HSE Moscow Institute of Electronics and Mathematics explain how AI technologies can improve an application, device, or system, and what engineering tasks are solved in the process.
‘The Peak of Stupidity’ and ‘The Valley of Despair’: HSE Economists Propose an Explanation for the Dunning–Kruger Effect
The Dunning–Kruger effect, which describes a sharp surge in self-confidence among beginners followed by an equally rapid decline as they gain experience, can be explained by the nature of the learning process and the acquisition of new knowledge. This conclusion was reached by Andrey Vorchik of the HSE Faculty of Economic Sciences together with independent researcher Murat Mamyshev. They developed a mathematical model of learning and demonstrated how subjective confidence is formed and changes as knowledge accumulates, as well as how teachers can reduce the ‘valley of despair’ experienced by learners.
Toffee and Risk: Scientists Discover Why People Who Crave Sweets Make More Impulsive Choices
Having a sweet tooth may be linked not only to eating habits but also to the way people make decisions. Researchers at HSE University have found that people with a preference for sweet foods tend to behave more impulsively—not because they want immediate rewards, but because they are less willing to tolerate uncertainty. These findings may help improve treatments for addiction. The study findings have been published in Frontiers in Psychology.
Advancing Collaboration: HSE Faculty of Computer Science and Harbin Institute of Technology Hold Joint Seminar
From July 13 to 16, 2026, the Faculty of Computer Science hosted the Russian–Sino Research Seminar on Machine Learning Applications, organised by the HSE Laboratory for Cloud and Mobile Technologies in partnership with the Harbin Institute of Technology (China). A delegation comprising seven students and three university representatives travelled to Moscow to take part in an intensive four-day programme.
Physicists Find a Way to Model Ion Parameters in Plasma in Seconds
Researchers from HSE University and the Moscow Institute of Physics and Technology (MIPT) have developed a set of simple analytical methods for calculating the properties of heavy ions in helium under the influence of a strong electric field. The new approach speeds up calculations of ion mobility and ion–molecule reaction rates by thousands of times while maintaining sufficient accuracy for plasma jet modelling. The findings have been published in the journal Physica Scripta.
A New Section on AI and a Prizewinning Paper: Early-Career HSE Researchers Take Part in IEEE EDM Conference
The 27th IEEE International Conference of Young Professionals in Electron Devices and Materials (EDM) has taken place in the Altai Republic. This year, researchers from HSE University presented the results of their research and were involved in organising a new section on artificial intelligence. A paper by HSE master’s student Rodion Sidorenko was awarded third place in the research paper competition at the conference.
Two Years of Growth or Decline: How to Choose an Investment Strategy
Economists from HSE University, together with colleagues from international universities, have analysed stock market movements over almost a century and proposed an investment strategy that could have delivered returns nearly twice as high as the market average. Their research suggests following a momentum strategy during periods of sustained market growth and switching to a value strategy after prolonged market declines. The study has been published in the Journal of Banking and Finance.
Researchers Reveal Link Between Attention and Communication Difficulties in Autism
Researchers at HSE University have examined how communication difficulties in children with autism are related to brain function. The findings show that not only language networks but also attention networks play an important role. The weaker the connections involved in maintaining focus and switching attention, the more pronounced communication difficulties were. The study has been published in European Child & Adolescent Psychiatry.
HSE Initiates Development of Ethical Standard for Anthropomorphic Robots
Beyond technological solutions, the development of anthropomorphic robotics also demands ethical ones. In July 2026, the HSE Institute for Robotics Systems hosted a foresight session dedicated to developing an Ethical Standard for Anthropomorphic Robotic Complexes. Representatives from businesses, government bodies, scientific organisations, and universities gathered to discuss key ethical and legal issues surrounding the development of anthropomorphic robotic complexes. The main outcome of the meeting was a draft of the Ethical Standard.