Team Name: PESTO
Problem: Predict S&P500 daily returns and implement a market timing strategy that buys and sells the index automatically based on a prediction model
Our Goal: Using a gradient boosting model and long-short term neural network, build a model for predicting the price of the S&P 500 Index, and use this prediction along with our strategy to buy and sell shares
https://www.youtube.com/watch?v=9zhrxE5PQgY&t=2415s This video explains the theories behind LSTM
https://www.coursera.org/learn/nlp-sequence-models Coursera that goes into the "how" of Neural Networks
https://machinelearningmastery.com/multivariate-time-series-forecasting-lstms-keras/ source code for LSTM
https://www.coursera.org/learn/nlp-sequence-models Coursera that goes into the "how" of Neural Networks
http://www.bioinf.jku.at/publications/older/2604.pdf Initial research paper that demonstrates what an LSTM does mathematically
StackNet GitHub repo: https://github.com/kaz-Anova/StackNet#what-is-stacknet
Bayesian Optimization GitHub repo: https://github.com/fmfn/BayesianOptimization
https://www.kaggle.com/tilii7/hyperparameter-grid-search-with-xgboost: Kaggle Optimization example for XGBoost
Big Dates:
Jan. 24 ddl for registration
January 27 - Our PCA is done, understand variables
January 31 - Test run first LSTM
February 15 - Prediction must be finished
February 22 - Strategy must be finished
Week of Mar. 2nd ddl for code test (prediction and strategy) and qualification test and proposal (creative)
Office Hours Friday 10-12, SH Cubicle 5432 X
Friday April 12 2020
Elia - Coursera courses on LSTMs, best optimization
Sergio - Putting the model in the cloud, adding users to aws
Anne - Looking up trading strategies for when markets go down
Patrick - Add Users to AWS, Look up Interactive Frameworks for website
Patrick Burke | Anqi Lin | Yiyi Xu | Sergio Zambrano |
---|---|---|---|
pwllmb |
anqilin11 |
eliaiye |
chilledapplesauce |