Model creation Source code
About this concert
Proprietary quantitative models and algorithmic trading strategies for long/short stock optimization models with specific risk and return parameters specified by the investor profile, using machine/deep learning, along with Q reinforcement learning agents.
Machine learning models using multivariate/logistic regression, lasso/ridge regression, linear/quadratic discriminant analysis, decision trees, K neighbors, Naive Bayes, random forest, support vector machine, Adaptiveboost, GradientBoost, XGB, and optimization portfolio to maximize return and minimize volatility for various investor risk profiles.
Deep learning modeling using recurrent neural networks, Tensorflow, nltk, sentiment analyzer, Keras LSTM and convolutional neural networks, in an attempt to predict the forecasted prices of specific asset classes across stocks, currencies, bonds, futures, ETFs and other derivatives.
Proprietary intraday long/short machine learning/deep algorithm using the Interactive Brokers API, IB_Insync Python library.