Volatility modeling - GARCH Parameter Estimation - Empirical challenges
3 important questions on Volatility modeling - GARCH Parameter Estimation - Empirical challenges
Name the 4 categories of empirical challenges with ML estimation for GARCH models
1. ML is a numerical approach, this can cause problems in small samples (can you explain why?)
2. GARCH (1,1) requires an initial value of σ1
3. ML function does not behave well
4. You have to perform robustness checks
2. GARCH (1,1) requires an initial value of σ1
3. ML function does not behave well
4. You have to perform robustness checks
One of the empirical challenges with ML estimation for GARCH models is that the GARCH(1,1) requires an initial value.
Name the 2 ways to get that value
1. Guess σ1 using EWMA or simple (historic) volatility.
2. Estimate σ1 as an additional parameter.
2. Estimate σ1 as an additional parameter.
Name 3 examples of ML not behaving well (in relation to GARCH estimation)
1. Multiple optima.
2. Narrow (global) optimum.
3. Non-unique solution.
2. Narrow (global) optimum.
3. Non-unique solution.
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
