Quantitative Finance Research Collection
A collection of technical projects exploring statistical modeling, deep learning, and model-selection techniques for financial time series.
Placeholder - Licensed financial analytics photography or actual research plots to be added
Overview
This research collection examines forecasting and classification methods for financial markets, with emphasis on model evaluation and comparison to baselines. Projects include regime-switching stochastic modeling for Bitcoin/S&P 500/Treasury yields, LSTM-based ETF return forecasting, SPY price-movement classification, and hyperparameter optimization techniques.
Methods & Technology
- Markov regime-switching models
- Stochastic volatility modeling
- LSTM time-series forecasting
- CNN classification for financial signals
- Hyperparameter optimization (Grid Search, Random Search, Bayesian)
- Ensemble learning methods
- Time-based train/validation/test splits
Limitations & Considerations
Educational and research projects. Does not constitute investment advice. Profitable trading performance not claimed. Transaction costs and real-market constraints not fully modeled. Proper time-based data splitting required for valid evaluation.
Discuss a Similar Project
Working on something similar? Let's talk about what you're building and see if we can help.