Chapter 10 (Dynamic regression models) Flashcards
1
Q
Dynamic regression models
A
allow the errors from a regression to contain autocorrelation
shown by ‘nt’
2
Q
Deterministic trend
A
yt = B0 + B1t + nt
nt is ARMA (no differencing)
3
Q
Stochastic trend
A
yt = B0 + B1t + nt
nt is ARIMA (with differencing)
4
Q
Problems with OLS
A
- Estimated coefficients are inconsistent
- Standard errors are incorrectly calculated
- Statistical inference is invalid
- AIC can be misleading
5
Q
Why use ARIMA structure on regression model?
A
- corrects the problems
- short term forecasts are more accurate
6
Q
Dynamic Harmonic Regression - Fourier terms with ARIMA errors
A
Benefits:
- captures seasonality with different frequencies
- controls smoothness of seasonal patterns by K
- short-term dynamics are easily handled
Con:
- seasonality is assumed to be fixed