On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,181 papers · 148 categories

Trend · papers per month

count % of corpus

9 results for “CSI300”

Long Short-Term Memory Networks for CSI300 Volatility Prediction with Baidu Search Volumeq-fin.ST

LSTM predicts CSI300 volatility using search volume data.

problem Accurate prediction of financial market volatility.

method Long Short-Term Memory (LSTM) neural network applied to Baidu search volume data.

result LSTM outperforms GARCH model in CSI300 volatility forecasting.

volatility prediction LSTM neural network financial markets Baidu search volume CSI300

2018-05-29abs ↗pdf ↗

Design of High-Frequency Trading Algorithm Based on Machine Learningq-fin.TR

New algorithm uses machine learning to predict high-frequency trading returns.

problem Improving prediction accuracy in high-frequency trading.

method Iterative optimization and activation functions in deep learning, combined with

VPIN

, GARCH, and SVM.

result The model significantly improved prediction of market liquidity and trading returns.

High-frequency trading Machine learning Volume-synchronized probability Support Vector Machine GARCH

2019-12-21abs ↗pdf ↗

Robust Log-Optimal Strategy with Reinforcement Learningq-fin.PM

RLOSRL improves portfolio management by combining RL with a robust strategy.

problem Portfolio optimization and resilience to estimation errors.

method Combining RL with RLOS, approximating objective function with Taylor expansion.

result RLOSRL outperforms traditional strategies in back tests.

portfolio management reinforcement learning log-optimal strategy reliability optimization

2018-05-01abs ↗pdf ↗

EFS: Evolutionary Factor Searching for Sparse Portfolio Optimization Using Large Language Modelsq-fin.PM

EFS uses LLMs to optimize sparse portfolios by evolving alpha factors.

problem Sparse portfolio optimization in dynamic market regimes.

method Evolutionary feedback loop with LLM-generated alpha factors.

result Significantly outperforms baselines in diverse datasets.

sparse portfolio optimization large language models alpha factors evolutionary search quantitative finance

2025-07-23abs ↗pdf ↗

Profitability of simple stationary technical trading rules with high-frequency data of Chinese Index Futuresq-fin.ST

Study on profitability of technical trading rules using high-frequency data of Chinese Index Futures.

problem Investigating the profitability of technical trading rules with high-frequency data of Chinese Index Futures.

method Converted MA, KDJ, and Bollinger bands into stationary processes and used ADF-test and SPA test to verify stationarity and assess trading rules' performance.

result Significant combinations of parameters for each indicator were found, but trading profits were eliminated with transaction costs included.

Chinese Index Futures Technical Trading Rules High-Frequency Data Stationary Processes ADF-test

2017-10-20abs ↗pdf ↗

TCGPN: Temporal-Correlation Graph Pre-trained Network for Stock Forecastingcs.LG

TCGPN improves stock forecasting by capturing temporal correlation patterns.

problem Stock forecasting with minimal periodicity and large node numbers.

method TCGPN uses Temporal-Correlation fusion encoder and pre-training methods to handle large datasets.

result TCGPN achieves state-of-the-art results on real stock market data.

stock forecasting temporal correlation graph neural networks pre-training memory efficiency

2024-07-26abs ↗pdf ↗

GRU-PFG: Extract Inter-Stock Correlation from Stock Factors with Graph Neural Networkq-fin.CP

GRU-PFG model extracts inter-stock correlations from stock factors using graph neural networks.

problem Limited effectiveness of models relying solely on stock factors for capturing stock correlations.

method Project stock factors into a graph and use graph neural networks to extract inter-stock correlations.

result Achieves better prediction results than models relying solely on stock factors and comparable to second category models.

stock prediction graph neural networks stock factors inter-stock correlation industry information

2024-11-28abs ↗pdf ↗

DiffsFormer: A Diffusion Transformer on Stock Factor Augmentationq-fin.ST

DiffsFormer uses AI-generated samples to improve stock forecasting accuracy.

problem Data scarcity in stock forecasting, including low signal-to-noise ratio and data homogeneity.

method DiffsFormer employs a Diffusion Model with a Transformer architecture to generate augmented stock factors.

result DiffsFormer achieves significant improvements in stock forecasting accuracy (7.2% and 27.8% relative improvements for CSI300 and CSI800 datasets, respectively).

stock forecasting data scarcity AI-generated samples diffusion model transformer architecture

2024-02-05abs ↗pdf ↗

From Knowing to Doing: A Memory-Controlled Benchmark for LLM Trading Agents on Stock Marketscs.AI

Benchmark evaluates LLM trading agents by masking identifiers to prevent memory leaks.

problem Evaluate LLM trading agents without relying on market memory or noise.

method Data-side masking protocol, Barra-style performance attribution framework.

result LLM agents' returns are largely explained by market and style exposure, not stock selection.

LLM trading benchmark memory masking performance attribution stock selection

2026-05-27abs ↗pdf ↗