Where the rainbow ends...
(2011)Department of Business Administration
- Abstract
- Purpose: The purpose of this thesis is to evaluate which model for forecasting the variance covariance matrix is the most accurate. This is important because the more accurate forecast the more correct pricing of derivatives with several underlying instruments.
Methodology: Using currency data of EUR/USD and GBP/USD five different variance covariance matrix forecasting models are used to price simulated currency rainbow option. The models are then evaluated by simulated trading between five agents using their own forecasting model which the one with the best forecasting model should have accumulated the higher profit. Statistical evaluation is also used to determine the most superior variance covariance matrix forecasting model. Results:... (More) - Purpose: The purpose of this thesis is to evaluate which model for forecasting the variance covariance matrix is the most accurate. This is important because the more accurate forecast the more correct pricing of derivatives with several underlying instruments.
Methodology: Using currency data of EUR/USD and GBP/USD five different variance covariance matrix forecasting models are used to price simulated currency rainbow option. The models are then evaluated by simulated trading between five agents using their own forecasting model which the one with the best forecasting model should have accumulated the higher profit. Statistical evaluation is also used to determine the most superior variance covariance matrix forecasting model. Results: The results shows that for the whole sample period the naïve forecasting model is the most superior when looking at the accumulated trading profit, followed by CCC-GARCH. According to the statistical loss evaluation DCC-GARCH is the superior model followed by CCC-GARCH. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/2426110
- author
- Näsholm, Armin and Norling, Nicklas
- supervisor
- organization
- year
- 2011
- type
- H1 - Master's Degree (One Year)
- subject
- keywords
- DCC-GARCH, CCC-GARCH, GARCH, Correlation forecasting, Rainbow option, Volatility forecasting, Management of enterprises, Företagsledning, management
- language
- Swedish
- id
- 2426110
- date added to LUP
- 2011-06-03 00:00:00
- date last changed
- 2012-11-12 11:27:35
@misc{2426110,
abstract = {{Purpose: The purpose of this thesis is to evaluate which model for forecasting the variance covariance matrix is the most accurate. This is important because the more accurate forecast the more correct pricing of derivatives with several underlying instruments.
Methodology: Using currency data of EUR/USD and GBP/USD five different variance covariance matrix forecasting models are used to price simulated currency rainbow option. The models are then evaluated by simulated trading between five agents using their own forecasting model which the one with the best forecasting model should have accumulated the higher profit. Statistical evaluation is also used to determine the most superior variance covariance matrix forecasting model. Results: The results shows that for the whole sample period the naïve forecasting model is the most superior when looking at the accumulated trading profit, followed by CCC-GARCH. According to the statistical loss evaluation DCC-GARCH is the superior model followed by CCC-GARCH.}},
author = {{Näsholm, Armin and Norling, Nicklas}},
language = {{swe}},
note = {{Student Paper}},
title = {{Where the rainbow ends...}},
year = {{2011}},
}