1. Introduction 1.1. Introduction

1.2. Download the academic publication

2. Tutorial on Correlation 2.1. Introduction 2.2. Stationarity, Correlation and the Pearsonr test 2.3. Linearly correlated series, Pearson versus Spearman 2.4. Intuition: Positive/Negative/Zero correlation. 3. Initial steps 3.1. Introduction 3.2. Data Preprocessing 3.3. Nonstationarity, KPSS and differencing 3.4. Find correlations and check their significance 3.5. Creating the data structures 4. The features extraction process 4.1. Introduction 4.2. Description 4.3. Theory 4.4. Overview of the code 4.5. Example: Features Extraction without scaling 4.6. Example: Features Extraction with scaling 4.7. Polynomials transform 5. Linear Regression 5.1. Introduction 5.2. Rolling predictions code 5.3. How to inverse the differencing operation 5.4. Going through the code using an example 5.5. Inversion of differencing (training predictions) 5.6. Inversion of differencing (test predictions) 5.7. Plot the training / test set predictions 5.7b. Test predictions using the math formula (lags approach) 5.7c. Test predictions using the math formula (polynomials transform) 5.8. Calculating the training errors 5.9. Calculating the test errors 5.10. Overfitting analysis 5.11. Naive benchmark test 5.12. Selection of models (lags approach) 5.13. Features extraction via polynomials transform 5.14. Selection of models (polynomials transform) 5.15. Generating the forecasts 5.16. Forecasts: Running the code and analysis 5.17. Revision of linear regression 6. Vector Autoregression (VAR) 6.1. Introduction 6.2. Rolling predictions 6.3. Plotting the predictions 6.3b. Test predictions using the math formula (lagged values) 6.4. Training and test predictions 6.5. Overfitting analysis 6.6. Selection of models 6.7. Comparing the test errors 6.8. Generating the forecasts 7. Shallow neural networks 7.1. Introduction 7.2. Rolling predictions 7.3. Training and test predictions 7.4. Training / test errors 7.5. Errors comparison 7.6. Overfitting analysis 7.7. Selection of models 7.8. Generating the forecasts 8. Deep neural networks 8.1. Introduction 8.2. Rolling predictions 8.3. Training and test predictions 8.4. Training / test errors 8.5. Overfitting analysis 8.6. Selection of models 8.7. Errors comparison 8.8. Generating the forecasts 9. Conclusions

9.1. Conclusions

9.2. Remarks