---
title: Stock Movement Prediction and Portfolio Optimization
url: https://www.ml-quant.com/papers/ssrn/5234375/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 5234375
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5234375
featured: 2025-04-30
citations: unknown
topic: Portfolio & Allocation
---


# Stock Movement Prediction and Portfolio Optimization

The study introduces a deep learning framework, MR2TNet, to predict NIFTY50 index stock price movements by considering multiple information sources.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5234375
- Identifier: SSRN 5234375
- Released: 2025-04-29
- First featured: Quant Letter No. 95 (2025-04-30): https://www.ml-quant.com/issues/2025-04-30/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Portfolio & Allocation

## Related

- [From Headlines to Holdings: Deep Learning for Smarter Portfolio Decisions](https://www.ml-quant.com/papers/arxiv/2509.24144/): A new deep learning framework for portfolio optimization, which combines LSTM networks, Graph Attention Networks, and financial news sentiment analysis, outperforms traditional benchmarks by directly learning portfolio weights, resulting in higher cumulative returns and Sharpe ratios.
- [Optimizing Portfolio Allocation](https://www.ml-quant.com/papers/ssrn/4534131/): The BlackLitterman model, BLEnd2End, uses deep learning to optimize portfolio allocation, outperforming mean-variance benchmarks and other traditional strategies.
- [Breaking the Dimensional Barrier: A Pontryagin-Guided Direct Policy Optimization for Continuous-Time Multi-Asset Portfolio Choice](https://www.ml-quant.com/papers/arxiv/2504.11116/): The article introduces the Pontryagin-Guided Direct Policy Optimization framework for solving large-scale portfolio optimization problems, capable of handling up to 50 assets and 10 state variables.
- [Causal Portfolio Optimization: Principles and Sensitivity-Based Solutions](https://www.ml-quant.com/papers/arxiv/2504.05743/): The article introduces a new risk management framework that uses Bayesian and neural networks for efficient portfolio optimization, based on Common Causal Manifolds.
- [A Scalable Gradient-Based Optimization Framework for Sparse Minimum-Variance Portfolio Selection](https://www.ml-quant.com/papers/arxiv/2505.10099/): The article introduces a quick and scalable gradient-based method for portfolio optimization, transforming the complex selection problem into a simpler task, with results comparable to commercial solvers and minimal error in portfolio variance.
- [Multiperiod Mean-Buffered Probability of Exceedance in Defined Contribution Portfolio Optimization](https://www.ml-quant.com/papers/arxiv/2505.22121/): The research investigates the effectiveness of buffered Probability of Exceedance strategies in long-term retirement planning through multi-period mean-risk portfolio optimization.