Technology Category
Machine Learning: Time-Series/Data StreamsData Mining & Knowledge Management: Anomaly/Outlier DetectionSearch and Optimization: Mixed Discrete/Continuous Search
Asset price movement prediction using empirical mode decomposition and Gaussian mixture models
🤖 AI Summary
This paper addresses the challenge of optimizing trading decisions under non-stationary financial time series. We propose a cross-asset price direction prediction framework integrating Empirical Mode Decomposition (EMD) and Gaussian Mixture Models (GMM). First, EMD extracts multi-scale intrinsic oscillatory components to enhance temporal discriminability. Second, GMM clusters market regimes to enable pattern-aware feature selection and synthetic data augmentation. Finally, a rolling-window time-series classifier—combining Random Forest and XGBoost—is trained and rigorously evaluated using time-series cross-validation. To our knowledge, this is the first work jointly leveraging EMD and GMM for cross-market (cryptocurrency and equity) pattern recognition and feature engineering. The approach significantly improves model robustness and profitability: EMD-derived features yield an average 18.3% increase in cumulative returns; after GMM-based feature filtering, over 70% of model–dataset configurations outperform the top decile of random strategies, thereby expanding the frontier of algorithm–data co-optimization.
Technology Category
Application Category
📝 Abstract
We investigated the use of Empirical Mode Decomposition (EMD) combined with Gaussian Mixture Models (GMM), feature engineering and machine learning algorithms to optimize trading decisions. We used five, two, and one year samples of hourly candle data for GameStop, Tesla, and XRP (Ripple) markets respectively. Applying a 15 hour rolling window for each market, we collected several features based on a linear model and other classical features to predict the next hour's movement. Subsequently, a GMM filtering approach was used to identify clusters among these markets. For each cluster, we applied the EMD algorithm to extract high, medium, low and trend components from each feature collected. A simple thresholding algorithm was applied to classify market movements based on the percentage change in each market's close price. We then evaluated the performance of various machine learning models, including Random Forests (RF) and XGBoost, in classifying market movements. A naive random selection of trading decisions was used as a benchmark, which assumed equal probabilities for each outcome, and a temporal cross-validation approach was used to test models on 40%, 30%, and 20% of the dataset. Our results indicate that transforming selected features using EMD improves performance, particularly for ensemble learning algorithms like Random Forest and XGBoost, as measured by accumulated profit. Finally, GMM filtering expanded the range of learning algorithm and data source combinations that outperformed the top percentile of the random baseline.
Problem
Research questions and friction points this paper is trying to address.
Predict asset price movements using EMD and GMM
Optimize trading decisions with machine learning
Improve performance via feature decomposition and clustering
Innovation
Methods, ideas, or system contributions that make the work stand out.
EMD decomposes features into multi-frequency components
GMM filters market data into clusters
Ensemble models classify movements with EMD-enhanced features
🔎 Similar Papers
No similar papers found.
