Simple is better. Empirical comparison of American option valuation methods
Technique for American options valuation, combining Least Squares Monte Carlo with Duan\’s model under the assumption that the volatility of the underlier can be described by GARCH(1, 1) process, has been confronted with simple binomial tree model. Results of comparison of model outcomes with market prices for ten different CBOE-traded stock options indicate that simple binomial model is superior to sophisticated GARCH-LSM method. The results hold regardless of option characteristics—“moneyness” ratio and time to maturity. Incorporating dividend in binomial model does not significantly alter the valuation outcomes. Detailed analysis shows also that for each of the methods pricing errors grow as the “moneyness” ratio decreases.
Year of publication: |
2012
|
---|---|
Authors: | Toporek, Katarzyna |
Published in: |
Ekonomia journal. - Wydział Nauk Ekonomicznych. - Vol. 29.2012, 29
|
Publisher: |
Wydział Nauk Ekonomicznych |
Subject: | Least Squares Monte Carlo | GARCH | option valuation | Duan model | LRNVR |
Saved in:
freely available
Saved in favorites
Similar items by subject
-
Volatility Components, Affine Restrictions and Non-Normal Innovations
Christoffersen, Peter, (2008)
-
Estimating and using GARCH models with VIX data for option valuation
Kanniainen, Juho, (2014)
-
GARCH Option Valuation: Theory and Evidence
Christoffersen, Peter, (2012)
- More ...