lateprogrammer/stock-prediction-models

Gathers machine learning and deep learning models for Stock forecasting, included trading bots and simulations

---

Stock-Prediction-Models, Gathers machine learning and deep learning models for Stock forecasting, included trading bots and simulations.

Table of contents

• Models
• Agents
• Data Explorations
• Simulations
• Results

Contents

Models

Stacking models

1. Deep Feed-forward Auto-Encoder Neural Network to reduce dimension + Deep Recurrent Neural Network + ARIMA + Extreme Boosting Gradient Regressor
2. Adaboost + Bagging + Extra Trees + Gradient Boosting + Random Forest + XGB

Deep-learning models

1. LSTM Recurrent Neural Network
2. Encoder-Decoder Feed-forward + LSTM Recurrent Neural Network
3. LSTM Bidirectional Neural Network
4. 2-Path LSTM Recurrent Neural Network
5. GRU Recurrent Neural Network
6. Encoder-Decoder Feed-forward + GRU Recurrent Neural Network
7. GRU Bidirectional Neural Network
8. 2-Path GRU Recurrent Neural Network
9. Vanilla Recurrent Neural Network
10. Encoder-Decoder Feed-forward + Vanilla Recurrent Neural Network
11. Vanilla Bidirectional Neural Network
12. 2-Path Vanilla Recurrent Neural Network
13. LSTM Sequence-to-Sequence Recurrent Neural Network
14. LSTM with Attention Recurrent Neural Network
15. LSTM Sequence-to-Sequence with Attention Recurrent Neural Network
16. LSTM Sequence-to-Sequence Bidirectional Recurrent Neural Network
17. LSTM Sequence-to-Sequence with Attention Bidirectional Recurrent Neural Network
18. LSTM with Attention Scaled-Dot Recurrent Neural Network
19. LSTM with Dilated Recurrent Neural Network
20. Only Attention Neural Network
21. Multihead Attention Neural Network
22. LSTM with Bahdanau Attention
23. LSTM with Luong Attention
24. LSTM with Bahdanau + Luong Attention
25. DNC Recurrent Neural Network
26. Residual LSTM Recurrent Neural Network

Agents

1. Simple signal rolling agent
2. Q-learning deep learning agent
3. Evolution-strategy agent
4. Evolution-strategy-Bayesian agent

Data Explorations

1. stock market study on TESLA stock, tesla-study.ipynb
2. fashion trending prediction with cross-validation, fashion-forecasting.ipynb
3. Bitcoin analysis with LSTM prediction, bitcoin-analysis-lstm.ipynb
4. Outliers study using K-means, SVM, and Gaussian on TESLA stock outliers.ipynb
5. Kijang Emas Bank Negara, kijang-emas-bank-negara.ipynb

Simulations

1. Stock market simulation using Monte Carlo, stock-forecasting-monte-carlo.ipynb
2. Stock market simulation using Monte Carlo Markov Chain Metropolis-Hasting, mcmc-stock-market.ipynb

Tensorflow-js

I code LSTM Recurrent Neural Network and Simple signal rolling agent inside Tensorflow JS, you can try it here, huseinhouse.com/stock-forecasting-js

Results

Results Agent

signal rolling agent, READ MORE

total gained 185.099850, total investment 1.850998 %

q-learning deep learning agent READ MORE

total gained -108.630036, total investment -1.086300 %

evolution strategy agent READ MORE

total gained 11037.529911, total investment 110.375299 %

evolution strategy with bayesian agent READ MORE

total gained 13295.469683, total investment 132.954697 %

Results signal prediction

LSTM Recurrent Neural Network

LSTM Bidirectional Neural Network

2-Path LSTM Recurrent Neural Network

Deep Feed-forward Auto-Encoder Neural Network to reduce dimension + Deep Recurrent Neural Network + ARIMA + Extreme Boosting Gradient Regressor

LSTM Sequence-to-Sequence Recurrent Neural Network

LSTM Sequence-to-Sequence with Attention Recurrent Neural Network

LSTM Sequence-to-Sequence with Attention Bidirectional Recurrent Neural Network

Encoder-Decoder Feed-forward + LSTM Recurrent Neural Network

Adaboost + Bagging + Extra Trees + Gradient Boosting + Random Forest + XGB