Educated Guess
1
Forecasting With Time Series Models
2
– Introduction to Forecasting
2.1
What Forecast Is and Is Not
2.2
A Brief History of the Study of Forecasting
2.3
Through the Lens of Causal Inference
2.4
Self Fulfilling Prophecy
2.5
Knowing the Unknown
2.6
Why We Can’t Get It Right
2.7
Minimizing the Risk of Getting It Wrong
2.8
Economists Do It With Models
2.9
Getting It Right for the Right Reasons
3
– Features of Time Series Data
3.1
Stochastic Process and Time Series
3.2
Stationarity and Ergodicity
3.3
White Noise Process
3.4
Autocorrelation
3.5
Autocorrelogram and It’s Forensic Features
3.6
Partial Autocorrelation
3.7
Transformations
3.8
Getting to the Root of It
4
– Generating and Evaluating Forecasts
4.1
Pseudo-Forecasting Routine
4.2
Forecast Assessment
4.2.1
Unbiasedness
4.2.2
Efficiency
4.2.3
No Autocorrelation
5
– Comparing Forecasts
5.1
The Need for the Forecast Evaluation
5.2
Relative Forecast Accuracy Tests
5.2.1
The Morgan-Granger-Newbold Test
5.2.2
The Diebold-Mariano Test
5.3
Forecasting Year-on-Year Monthly Inflation 12-steps-ahead
6
– Combining Forecasts
6.1
Benefits of Forecast Combination
6.2
Optimal Weights for Forecast Combination
6.3
Forecast Encompassing
7
– Trends
7.1
Trends in the Data
7.2
Spurious Relationships
7.2.1
Deterministic Trends
7.2.2
Stochastic Trends
7.3
Modeling
7.3.1
Trends in mortgage rates
7.4
Forecasting
8
– Seasonality
8.1
Seasonal Fluctuations in the Data
8.2
Modeling
8.2.1
Seasonal dummy variables
8.2.2
Seasonal harmonic variables
8.3
Forecasting
9
– Autoregression
9.1
Stochastic Cycles
9.2
Modeling
9.2.1
First-order autoregression
9.2.2
Unit Roots and Non-stationarity
9.3
Forecasting
9.3.1
Iterative Method of Multistep Forecasting
9.3.2
Direct Method of Multistep Forecasting
10
– Vector Autoregression
10.1
Dynamic Feedbacks Among Economic Variables
10.2
Modeling
10.2.1
In-Sample Granger Causality
10.3
Forecasting
10.3.1
One-step-ahead forecasts
10.3.2
Multi-step-ahead forecasts
10.3.3
Out-of-Sample Granger Causality
11
– Threshold Autoregression
11.1
Regime-Dependent Nonlinearity
11.2
Modeling
11.3
Forecasting
11.3.1
Skeleton Extrapolation
11.3.2
Analytical Method
11.3.3
Numerical Method: Bootstrap Resampling
Forecasting Using R
Tutorial 1: Introduction to R
Base R and matrix manipulations
Estimating parameters using OLS
Tutorial 2: Data Management and Visualisation
Tutorial 3: Forecasting Methods and Routines
Tutorial 4: Trends
Tutorial 5: Seasonality
Tutorial 6: Autoregression
Tutorial 7: Vector Autoregression
Tutorial 8: Threshold Autoregression
Tutorial 9: Comparing Forecasts
Tutorial 10: Combining Forecasts
References
Educated Guess
Tutorial 5: Seasonality