Useful Links
Computer Science
Data Science
Time Series Analysis and Forecasting
1. Introduction to Time Series Data
2. Mathematical Foundations
3. Fundamental Concepts
4. Data Preprocessing and Exploration
5. Classical Forecasting Models
6. Advanced Statistical Models
7. Machine Learning for Time Series
8. Model Evaluation and Validation
9. Advanced Topics and Applications
Model Evaluation and Validation
Forecasting Horizons
One-Step-Ahead Forecasting
Point forecasts
Interval forecasts
Multi-Step-Ahead Forecasting
Direct strategy
Recursive strategy
DirRec strategy
MIMO strategy
Performance Metrics
Scale-Dependent Errors
Mean Error
Mean Absolute Error
Mean Squared Error
Root Mean Squared Error
Percentage Errors
Mean Percentage Error
Mean Absolute Percentage Error
Symmetric Mean Absolute Percentage Error
Scale-Independent Errors
Mean Absolute Scaled Error
Relative Root Mean Squared Error
Distributional Accuracy
Quantile loss
Continuous ranked probability score
Energy score
Directional Accuracy
Hit rate
Directional symmetry
Validation Strategies
Hold-Out Validation
Train-test split
Temporal ordering preservation
Cross-Validation Techniques
Time series cross-validation
Rolling forecast origin
Expanding window validation
Sliding window validation
Walk-Forward Analysis
Out-of-sample testing
Retraining frequency
Model updating strategies
Residual Analysis
Residual Properties
Zero mean assumption
Constant variance
Independence assumption
Normality assumption
Diagnostic Tests
Ljung-Box test
Durbin-Watson test
Breusch-Godfrey test
ARCH test
Visual Diagnostics
Residual plots
Q-Q plots
Histogram analysis
ACF of residuals
Model Selection
Information Criteria
Akaike Information Criterion
Bayesian Information Criterion
Hannan-Quinn Criterion
Cross-Validation Scores
Time series CV error
Rolling window performance
Forecast Accuracy Comparison
Statistical significance tests
Diebold-Mariano test
Model confidence sets
Previous
7. Machine Learning for Time Series
Go to top
Next
9. Advanced Topics and Applications