Narrow search
Showing 1 - 10 of 7,653
Cover Image
Asymptotic Inference in Multiple-Threshold Nonlinear Time Series Models
Li, Dong; Ling, Shiqing; Zakoian, Jean-Michel - Centre de Recherche en Économie et Statistique … - 2013
This paper investigates a class of multiple-threshold models, called Multiple Threshold Double AR (MTDAR) models. A sufficient condition is obtained for the existence and uniqueness of a strictly stationary and ergodic solution to the first-order MTDAR model. We study the Quasi-Maximum...
Persistent link: https://www.econbiz.de/10010747015
Cover Image
GARCH models without positivity constraints: Exponential or log GARCH?
Francq, Christian; Wintenberger, Olivier; Zakoïan, … - In: Journal of Econometrics 177 (2013) 1, pp. 34-46
This paper provides a probabilistic and statistical comparison of the log-GARCH and EGARCH models, which both rely on …, existence of moments, tails) of the EGARCH model, which are already known, with those of an asymmetric version of the log-GARCH …. The quasi-maximum likelihood estimation of the log-GARCH parameters is shown to be strongly consistent and asymptotically …
Persistent link: https://www.econbiz.de/10011052251
Cover Image
GARCH models without positivity constraints: Exponential or Log GARCH?
Zakoïan, Jean-Michel; Wintenberger, Olivier; Francq, … - Université Paris-Dauphine (Paris IX) - 2013
This paper provides a probabilistic and statistical comparison of the log-GARCH and EGARCH models, which both rely on …, existence of moments, tails) of the EGARCH model, which are already known, with those of an asymmetric version of the log-GARCH …. The quasi-maximum likelihood estimation of the log-GARCH parameters is shown to be strongly consistent and asymptotically …
Persistent link: https://www.econbiz.de/10011072512
Cover Image
An ARCH model without intercept
Hafner, Christian M.; Preminger, Arie - In: Economics Letters 129 (2015) C, pp. 13-17
While theory of autoregressive conditional heteroskedasticity (ARCH) models is well understood for strictly stationary processes, some recent interest has focused on the nonstationary case. In the classical model including a positive intercept parameter, the volatility process diverges to...
Persistent link: https://www.econbiz.de/10011263447
Cover Image
Forecasting chaotic systems : the role of local Lyapunov exponents.
Guegan, Dominique; Leroux, Justin - Centre d'Économie de la Sorbonne, Université Paris 1 … - 2008
We propose a novel methodology for forecasting chaotic systems which is based on the nearest-neighbor predictor and improves upon it by incorporating local Lyapunov exponents to correct for its inevitable bias. Using simulated data, we show that gains in prediction accuracy can be substantial....
Persistent link: https://www.econbiz.de/10005670880
Cover Image
The North American natural gas liquids markets are chaotic
Serletis, Apostolos; Gogas, Periklis - Volkswirtschaftliche Fakultät, … - 1999
In this paper we test for deterministic chaos (i.e., nonlinear deterministic processes which look random) in seven Mont Belview, Texas hydrocarbon markets, using monthly data from 1985:1 to 1996:12--the markets are those of ethane, propane, normal butane, iso-butane, naptha, crude oil, and...
Persistent link: https://www.econbiz.de/10005789499
Cover Image
Random walk or chaos: A formal test on the Lyapunov exponent
Park, Joon Y.; Whang, Yoon-Jae - In: Journal of Econometrics 169 (2012) 1, pp. 61-74
A formal test on the Lyapunov exponent is developed to distinguish a random walk model from a chaotic system, which is based on the Nadaraya–Watson kernel estimator of the Lyapunov exponent. The asymptotic null distribution of our test statistic is free of nuisance parameter, and simply given...
Persistent link: https://www.econbiz.de/10010577524
Cover Image
Predicting chaos with Lyapunov exponents : zero plays no role in forecasting chaotic systems.
Guegan, Dominique; Leroux, Justin - Centre d'Économie de la Sorbonne, Université Paris 1 … - 2010
We propose a nouvel methodology for forecasting chaotic systems which uses information on local Lyapunov exponents (LLEs) to improve upon existing predictors by correcting for their inevitable bias. Using simulations of the Rössler, Lorenz and Chua attractors, we find that accuracy gains can be...
Persistent link: https://www.econbiz.de/10008622043
Cover Image
Premières preuves empiriques de chaos dans les ventes de biens à la mode - First empirical evidence of chaos in the sales of fashion goods
Bonache, Adrien; Moris, Karen - Université de Bourgogne - CREGO EA7317 Centre de … - 2011
(VF)L’objectif de cet article est de tester la présence de chaos dans les ventes de biens à la mode. Après avoir brièvement rappelé le débat théorique autour de cette hypothèse, nous présentons les résultats. Ces résultats renforcent la plausibilité de l’existence d’un...
Persistent link: https://www.econbiz.de/10009146898
Cover Image
Etude de la dynamique non-linéaire des rentabilités de la bourse de Casablanca
RIANE, Nizare - Volkswirtschaftliche Fakultät, … - 2014
The preponderance of the linear approach in the stock market modeling is the result of the Frisch-Slutsky paradigm which implies that the market can only converge to an equilibrium point or diverge, according to a monotonic or oscillatory trajectory. Moreover, this description of reality is...
Persistent link: https://www.econbiz.de/10011156979