Eric991005 / TFSTL
This repository provides a PyTorch implementation of "Trade Forecasting via Efficient Multi-commodity STL Decomposition based Neural Networks," a model enhancing global trade forecasting accuracy through sophisticated STL decomposition and neural network strategies.
☆13Updated 6 months ago
Related projects ⓘ
Alternatives and complementary repositories for TFSTL
- ☆10Updated 2 years ago
- 复现AAAI-21-STHAN-SR☆10Updated last year
- StockFormer: A Swing Trading Strategy Based on STL Decomposition and Self-Attention Networks☆87Updated 5 months ago
- A study on volume-price factor stock selection model based on wavelet transform and multitask self-attention network☆40Updated 4 months ago
- Reproduce AAAI22-FactorVAE☆55Updated last year
- Stock factor mining with CNN and GRU.☆40Updated last year
- The official implementation of the paper "MTMD: Multi-Scale Temporal Memory Learning and Efficient Debiasing Framework for Stock Trend Fo…☆28Updated 9 months ago
- Stock risk premium prediction via FM/ EXT/ GBDT/ XGB/LBGM. Mengxuan Chen's graduation thesis at WHU.☆12Updated 4 years ago
- 多因子选股框架☆21Updated 3 years ago
- Apply machine learning algorithms in the financial market. Ensemble Model, including XGBoost, LightGBM, CNN, ResNet and LSTM.☆10Updated 2 years ago
- my first factor-stock-selecting backtest function☆18Updated 4 years ago
- Enhance the gplearn package to support precise three-dimensional structured dimension genetic programming (GP), with a particular focus …☆28Updated 2 months ago
- 通过遗传算法、强化学习来自动选择高频因子☆23Updated last year
- ☆49Updated 2 years ago
- Stock Price Prediction with PCA and LSTM☆14Updated 3 years ago
- Backtest Framework designed by YuminQuant&Yumin.☆11Updated 2 months ago
- 基于基因表达式规划算法的因子挖掘☆24Updated 3 years ago
- ☆44Updated 3 years ago
- Unofficial PyTorch implementation of FactorVAE☆15Updated last year
- Transformer and MultiTransformer layers for stock volatility forecasting purposes☆63Updated 3 years ago
- Multi Task Learning Time Series Momentum☆14Updated 5 months ago
- ARIMA & GARCH models for stock price prediction☆17Updated 4 years ago
- ☆10Updated 3 years ago
- High frequency factors based on order and trade data.☆32Updated 10 months ago
- 雪球结构产品定价☆22Updated last year
- 改写了gplearn源码,原有的gplearn会把数据转为numpy,丢失了datetime和stockcode的原始信息。很难做截面的因子ic、ir分析,所以改动了相应的源码,使之可以做因子的截面ic分析。另外增加了时序函数和并行化框架ray的支持。☆16Updated 7 months ago
- 复现华泰证券《强化学习初探与DQN择时》研报中的DQN模型与效果