Pasar al contenido principal

PREDICCION DE SERIES TEMPORALES NO ESTACIONARIAS DE GRANDES DIMENSIONES

PID2019-108079GB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2019
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD CARLOS III DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Integrated nested Laplace approximations for threshold stochastic volatility models

  • Zea Bermúdez, Patricia de
  • Marín, Juan Miguel
  • Rue, Havard
  • Veiga, Helena
The aim is to implement the integrated nested Laplace approximations (INLA), known to be very fast and efficient, for estimating the parameters of the threshold stochastic volatility (TSV) model. INLA replaces Markov chain Monte Carlo (MCMC) simulations with accurate deterministic approximations. Weakly informative proper priors are used, as well as Penalizing Complexity (PC) priors. The simulation results favor the use of PC priors, specially when the sample size varies from small to moderate. For these sample sizes, PC priors provide more accurate estimates of the model parameters. However, as sample size increases, both types of priors lead to similar estimates of the parameters. The estimation method is applied to six series of returns, including stock market, commodity and cryptocurrency returns, and its performance is assessed, by means of in-sample and out-of-sample approaches; the forecasting of one-day-ahead volatilities is also carried out. The empirical results support that the TSV is the model that generally fits the best to the series of returns and most of the times ranks the first in terms of forecasting one-day-ahead volatility, when compared to the symmetric stochastic volatility model., The first author was partially financed by national funds through FCT - Fundação para a Ciência e a Tecnologia under the projects PTDC/MAT-STA/28649/2017 and UIDB/00006/2020. The second author acknowledges financial support from the Spanish Ministry of Science, Innovation and Universities, research project MTM2015-69638-R. The fourth author acknowledges financial support from the Spanish Ministry of Science, Innovation and Universities, research project PGC2018-096977-B-l00, from the Agencia Estatal de Investigación PID2019-108079GB-C21/AIE/10.13039/501100011033 and from Fundação para a Ciência e a Tecnologia, grant UIDB/00315/2020.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

An experimental analysis of contagion in financial markets

  • Peeters, Ronald
  • Veiga, Helena
  • Vorsatz , Marc
In an experimental market, we study how information about the dividend of an asset, which is available to some traders, is absorbed in the asset's price when all traders also have access to the prices of another different asset. We consider two treatments: in one, the dividends of the two assets are independent; in the other, the dividend of the own asset correlates positively with the dividend of the other asset. Since there is no aggregate uncertainty in the own market, the other dividend should not affect own prices according to the rational expectations equilibrium. Consistent with a prior information perspective, we find that (a) own prices are increasing in the other dividend if and only if dividends are correlated, and (b) correlated dividends can worsen information dissemination when the own dividend is low, and the other dividend is high. These findings imply that linkages between markets, both via fundamentals and via observability of market prices, can cause financial contagion even if there are corrective market forces at play (superior private information in the own market)., We acknowledge financial support from the Spanish Ministry of Science, Innovation and Universities, research projects PID2019-108079GB-C21 (MINECO/FEDER) and PID2021-122919NB-I00, from the Fundação para a Ciência e a Tecnologia, grant UIDB/00315/2020, and from the Observatorio Social de "La Caixa", project "Financial markets contagion: An experimental analysis" (2019-2020).
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Shock-triggered asymmetric response stochastic volatility

  • Marín, Juan Miguel
  • Veiga, Helena
We propose a novel asymmetric stochastic volatility model (STAR-SV) in which the leverage parameter adjusts to the magnitude of past shocks. This flexible specification captures both the leverage effects and their propagation more effectively than standard asymmetric volatility models. To estimate the STAR-SV parameters, we implement a data cloning algorithm that approximates the maximum likelihood estimates and their asymptotic variances. In finite-sample simulations, data cloning consistently leads to reliable estimates and small standard errors. Empirically, we fit the model to Bitcoin, Nasdaq, and S&P 500 returns and evaluate 1- and 10-day volatility forecasts using unconditional and conditional tests of predictive ability. STAR-SV using data cloning proves to be the most adequate forecaster, outperforming the most stringent confidence thresholds and weakly dominating in several variance regimes. Finally, we show the performance of the model in predicting the 99% value-at-risk. STAR-SV, using data cloning, seems to respond quickly to volatility spikes and passes backtests for both time horizons., This work was supported by Agencia Estatal de Investigación (PID2020-115598RB-I00, PID2019-108079GB-C21/AIE/10.13039/501100011033) and Fundação para a Ciência e a Tecnologia (UIDB/00315/2020).