Deep momentum networks with market trend dynamics
1Graduate School of Data Science, Chonnam National University, Gwangju, Republic of Korea.
Plos One
|September 2, 2025
Summary
This study enhances time-series momentum (TSMOM) trading strategies by integrating market trend data. The improved model, using Long Short-Term Memory (LSTM) and MTDP scores, shows better performance during market shifts.
Area of Science:
- Quantitative Finance
- Machine Learning in Finance
- Algorithmic Trading
Background:
- Time-series momentum (TSMOM) strategies rely on return trend persistence.
- Long Short-Term Memory (LSTM) networks can improve TSMOM but struggle with market trend changes.
Purpose of the Study:
- To enhance TSMOM performance during significant shifts in market trends.
- To develop a model that incorporates market dynamics for improved trading decisions.
Main Methods:
- Combined short- and long-term signals into a market-state representation using supervised learning.
- Generated market trend dynamic prediction (MTDP) scores via extreme gradient boosting (XGBoost).
- Applied MTDP scores within an LSTM-based trading strategy.
Main Results:
- Backtesting on 99 futures (1995-2021) showed MTDP score integration improved the Sharpe ratio.
- An 8-week momentum window excelled in stable periods (1995-2019).
- A 20-week window demonstrated superior performance and faster recovery during extreme downturns (e.g., COVID-19).
Conclusions:
- Integrating market-state information (MTDP scores) effectively enhances TSMOM strategies.
- Dynamically adjusting momentum look-back windows is crucial for consistent profitability in volatile markets.
Related Concept Videos
01:07End Point Prediction: Gran Plot
580
A Gran plot is used to predict the equivalence volume or endpoint of a potentiometric or acid-base titration without reaching the endpoint. Typically, titration data is collected as a function of the titrant's volume up to a point less than the equivalence volume and then transformed into a linear format. The straight line is extended to the x-axis, indicating the necessary titrant volume to achieve the equivalence point.
For potentiometric titration, the Gran plot is created by plotting...
For potentiometric titration, the Gran plot is created by plotting...
580
02:20Dynamic Equilibrium
53.3K
A reversible chemical reaction represents a chemical process that proceeds in both forward (left to right) and reverse (right to left) directions. When the rates of the forward and reverse reactions are equal, the concentrations of the reactant and product species remain constant over time and the system is at equilibrium. A special double arrow is used to emphasize the reversible nature of the reaction. The relative concentrations of reactants and products in equilibrium systems vary greatly;...
53.3K
01:24Sequence Networks of Rotating Machines
140
A Y-connected synchronous generator, grounded through a neutral impedance, is designed to produce balanced internal phase voltages with only positive-sequence components. The generator's sequence networks include a source voltage that is exclusively in the positive-sequence network. The sequence components of line-to-ground voltages at the generator terminals illustrate this configuration.
Zero-sequence current induces a voltage drop across the generator's neutral impedance and other...
Zero-sequence current induces a voltage drop across the generator's neutral impedance and other...
140
01:27Energy Line and Hydraulic Gradient Line
1.4K
Based on Bernoulli's equation, the energy line (EL) and hydraulic grade line (HGL) provide graphical representations of energy distribution in a fluid flow system. For steady, incompressible, inviscid flows, Bernoulli's equation is expressed as:
1.4K
01:10Central Tendency: Analysis
212
Measures of central tendency are tools used in biostatistics to identify the average or center of a dataset. They offer a single representative value for understanding and summarizing data distribution.
The mean is one such measure, calculated by totaling all values in a dataset and dividing by the number of values. For instance, the mean blood pressure reading (120, 130, 140, 150) would be 135. However, the mean can be affected by extreme values or outliers.
The median, another measure,...
The mean is one such measure, calculated by totaling all values in a dataset and dividing by the number of values. For instance, the mean blood pressure reading (120, 130, 140, 150) would be 135. However, the mean can be affected by extreme values or outliers.
The median, another measure,...
212
01:52Regression Toward the Mean
6.5K
Regression toward the mean (“RTM”) is a phenomenon in which extremely high or low values—for example, and individual’s blood pressure at a particular moment—appear closer to a group’s average upon remeasuring. Although this statistical peculiarity is the result of random error and chance, it has been problematic across various medical, scientific, financial and psychological applications. In particular, RTM, if not taken into account, can interfere when...
6.5K