Template-Type: ReDIF-Paper 1.0
Series: Tinbergen Institute Discussion Papers
Creation-Date: 2026-10-05
Number: 26-082/II
Author-Name: Fernando Roders
Author-Workplace-Name: a.s.r. Asset Management
Author-Name: Siem Jan Koopman
Author-Workplace-Name: Vrije Universiteit Amsterdam
Author-Name: Karim Moussa
Author-Workplace-Name: Vrije Universiteit Amsterdam
Title: Representation over Rents: Stochastic Volatility Models with Stable Disturbances: Estimation via Linearization
Abstract: In this study, we propose to estimate the parameters of stochastic volatility models with stable disturbances using methods based on linearization. Specifically, we obtain a linear state space form of the model by taking the logarithm of the squared financial returns, which, together with analytical expressions for logarithmic and fractional lower order moments of the stable distribution, forms the basis of the proposed quasi-maximum likelihood and generalized method of moments estimators. The methods are straightforward to implement and do not require simulation. We provide details of the implementation and study the finite-sample performance of the two estimation methods in a Monte Carlo study. The empirical relevance is illustrated by studying model-implied volatility estimates from time series of daily returns from stock indices, exchange rates, and cryptocurrencies.
Classification-JEL: C32, C58, G15
Keywords: 
File-URL: https://papers.tinbergen.nl/26082.pdf
File-Format: application/pdf
File-Size: 1.111.227 bytes
Handle: RePEc:tin:wpaper:20260082