Last Update: November, 2017
1. Course Objective
Learn Machine Trading Analysis main topics using Python programming language® in this practical course for all knowledge levels. Feel free to take a look at Course Curriculum.
2. Skills Learned
At the end of this course you will know how to:
- Create target and predictor features (supervised regression machine learning task).
- Select relevant predictor features subset through univariate filter methods (false discovery rate, family-wise error rate), deterministic wrapper method (recursive feature elimination) and embedded method (least absolute shrinkage and selection operator LASSO).
- Extract predictor features transformations (principal component analysis PCA).
- Train algorithms such as ensemble methods (gradient boosting machine GBM), maximum margin methods (radial basis function support vector machine RBF-SVM), multi-layer perceptron methods (artificial neural networks ANN) and estimate their optimal parameters through time series cross-validation.
- Test algorithms forecasting accuracy (mean absolute error, mean squared error, root mean squared error, mean absolute percentage error).
- Outline highest forecasting accuracy algorithms long-only machine trading strategies (target feature prediction centerline cross-over trading signals, strategy positions).
- Assess long-only machine trading strategies performance (annualized return, standard deviation and Sharpe ratio, cumulative returns chart) and compare them with buy and hold benchmark.
3. Typical Student
This course is ideal for you as:
- Undergraduate or postgraduate who wants to learn about the subject.
- Finance professional or academic researcher who wishes to deepen your knowledge in computational finance.
- Experienced investor who desires to research machine trading strategies.
- This course is NOT about “get rich quick” trading systems or magic formulas.