Missing observations in observation-driven time series models
Autor: | Siem Jan Koopman, Paolo Gorgi, Francisco Blasques |
---|---|
Přispěvatelé: | Econometrics and Data Science, Tinbergen Institute |
Rok vydání: | 2021 |
Předmět: |
Economics and Econometrics
Computer science Missing data Monte Carlo method Inference Indirect Inference 01 natural sciences Formal proof Indirect inference 010104 statistics & probability symbols.namesake Empirical research 0502 economics and business Econometrics 0101 mathematics Observation-driven models 050205 econometrics Series (mathematics) Applied Mathematics 05 social sciences Volatility symbols Consistency Volatility (finance) Gaussian network model |
Zdroj: | Journal of Econometrics, 221(2), 542-568. Elsevier BV Blasques, F, Gorgi, P & Koopman, S J 2021, ' Missing observations in observation-driven time series models ', Journal of Econometrics, vol. 221, no. 2, pp. 542-568 . https://doi.org/10.1016/j.jeconom.2020.07.043 |
ISSN: | 0304-4076 |
Popis: | We argue that existing methods for the treatment of missing observations in time-varying parameter observation-driven models lead to inconsistent inference. We provide a formal proof of this inconsistency for a Gaussian model with time-varying mean. A Monte Carlo simulation study supports this theoretical result and illustrates how the inconsistency problem extends to score-driven and, more generally, to observation-driven models, which include well-known models for conditional volatility. To overcome the problem of inconsistent inference, we propose a novel estimation procedure based on indirect inference. This easy-to-implement method delivers consistent inference. The asymptotic properties of the new method are formally derived. Our proposed estimation procedure shows a promising performance in a Monte Carlo simulation exercise as well as in an empirical study concerning the measurement of conditional volatility from financial returns data. |
Databáze: | OpenAIRE |
Externí odkaz: |