VALUE-AT-RISK PREDICTION FOR ARCH(1) AND SVAR(1) MODELS
<br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> Value-at-Risk is a risk measure to predict maximum loss of assets. In this thesis, we are c...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/16760 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | <br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
Value-at-Risk is a risk measure to predict maximum loss of assets. In this thesis, we are concerned with VaR prediction using volatility models, namely Autoregressive Conditional Heteroscedastic (ARCH) and Stochastic Volatility Autoregressive (SVAR). Both models have differences in estimating parameters and determining VaR prediction. Maximum likelihood method is used for estimating parameters of ARCH(1) model, while SVAR(1) model uses maximum likelihood-efficient important sampling method. Then, we will do backtesting to evaluate VaR prediction using coverage probability and correct VaR. Simulations have carried out to estimate parameters and predict the VaR for ARCH(1) and SVAR(1) models. |
|---|