ARCH MODEL: VALUE-AT-RISK AND COPULA-BASED PREDICTION
Modelling a risk to a stochastic model is commonly used in risk management. One of stochastic models which can capture heteroscedacticity property is Autoregressive Conditional Heteroscedastic (ARCH) model. In this thesis, some alternative representation of ARCH model are constructed. These new...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/32157 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Modelling a risk to a stochastic model is commonly used in risk management. One
of stochastic models which can capture heteroscedacticity property is
Autoregressive Conditional Heteroscedastic (ARCH) model. In this thesis, some
alternative representation of ARCH model are constructed. These new
representation giving us a new perspective about ARCH model. Value-at-Risk is a
risk measure that still can be developed. This thesis will study some alternative
calculation of VaR, using correlation measure between two random variables. As a
innovation, we will try to calculating VaR using Copula. Copula is a distribution
function model for two or more random variables. There is many kind of Copula
model. Then, Akaike Information Criterion will be used to selecting the best
Copula model to data. Alternative calculation of VaR then applied to two stock
return. As the result, correlation-based VaR giving a more effective and accurate
VaR prediction. |
|---|