Jean-Marc Bardet
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jean-Marc Bardet has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Paul Doukhan
AGM Mathematics UMR 8080 CY Cergy Paris University 2 Bd. Adolphe Chauvin 95000 Cergy-Pontoise AGM
5 shared publications - Béchir Dola
SAMM Université Panthéon-Sorbonne (Paris I) 90 rue de Tolbiac 75013 Paris FRANCE
3 shared publications Imen Kammoun
SAMM, Université Paris 1 , 90 rue de Tolbiac, Paris , France
3 shared publications- León R. José Rafael
Universidad Central de Venezuela
3 shared publications Véronique Billat
2 shared publications- Abdellatif Guenaizi
SAMM Université Panthéon-Sorbonne (Paris I) 90 rue de Tolbiac 75013 Paris FRANCE
1 shared publication
Latest Publications
- 19 February 2026
- Published by Wiley in Journal of Time Series Analysis
Stationarity and Goodness‐of‐Fit Tests for Locally Stationary Time Series
This paper considers the trajectory of a time series with time‐varying coefficients and proposes to test the adequacy of these parameters at a finite and fixed number of instants of the trajectory. For this purpose, a Wald test is constructed from point estimates of the parameters obtained by minimization of a kernel contrast. This can take the form of a localized near‐maximum likelihood estimator for ARMA or GARCH processes, or a localized least squares estimator for a GLARCH process, but many other time‐varying time series such as AR, ARCH, ARMA‐GARCH, APARCH,…, could be considered. Above all, this allows the introduction of a new stationarity test for these processes, whose very good numerical performance has been demonstrated by numerical experiments.
- 19 February 2026
- Published by Wiley in Journal of Time Series Analysis
Stationarity and Goodness‐of‐Fit Tests for Locally Stationary Time Series
This paper considers the trajectory of a time series with time‐varying coefficients and proposes to test the adequacy of these parameters at a finite and fixed number of instants of the trajectory. For this purpose, a Wald test is constructed from point estimates of the parameters obtained by minimization of a kernel contrast. This can take the form of a localized near‐maximum likelihood estimator for ARMA or GARCH processes, or a localized least squares estimator for a GLARCH process, but many other time‐varying time series such as AR, ARCH, ARMA‐GARCH, APARCH,…, could be considered. Above all, this allows the introduction of a new stationarity test for these processes, whose very good numerical performance has been demonstrated by numerical experiments.
- Open Access
A new nonparametric estimation method for trended locally stationary processes
This paper deals with locally stationary time series whose model is known (typically time-varying (tv)-ARMA, tv-GARCH, tv-ARCH(∞),...) observed with a nonparametric additive trend. Using a kernel-localised Gaussian quasi-maximum likelihood estimator, we are able to prove central limit theorems for different point values of the various functions to be estimated. In particular, we show that in the case of tv-GARCH or tv-ARCH(∞) processes, the trend estimator verifies asymptotic normality with an asymptotic variance smaller than that of the usual Nadaraya-Watson-type trend estimator. Numerical experiments illustrate the performance of the estimators on simulated data, and a meteorological example shows the value of coupling trend and parameter estimation for a locally stationary process.
Open AccessA new nonparametric estimation method for trended locally stationary processes
This paper deals with locally stationary time series whose model is known (typically time-varying (tv)-ARMA, tv-GARCH, tv-ARCH(∞),...) observed with a nonparametric additive trend. Using a kernel-localised Gaussian quasi-maximum likelihood estimator, we are able to prove central limit theorems for different point values of the various functions to be estimated. In particular, we show that in the case of tv-GARCH or tv-ARCH(∞) processes, the trend estimator verifies asymptotic normality with an asymptotic variance smaller than that of the usual Nadaraya-Watson-type trend estimator. Numerical experiments illustrate the performance of the estimators on simulated data, and a meteorological example shows the value of coupling trend and parameter estimation for a locally stationary process.
- 17 November 2024
- Published by Taylor & Francis in Statistics
- Vol. 59 (1) , 113-135
An irregularly spaced ARMA(1,1) model and an application to contamination data
- 17 November 2024
- Published by Taylor & Francis in Statistics
- Vol. 59 (1) , 113-135
An irregularly spaced ARMA(1,1) model and an application to contamination data
- 6 August 2024
- Published by Elsevier in Journal of Statistical Planning and Inference
- Vol. 235
A criterion for estimating the largest linear homoscedastic zone in Gaussian data
- 6 August 2024
- Published by Elsevier in Journal of Statistical Planning and Inference
- Vol. 235
A criterion for estimating the largest linear homoscedastic zone in Gaussian data
- 13 July 2024
- Published by Springer Nature in Statistical Inference for Stochastic Processes
- Vol. 27 (3) , 457-483
Quasi-maximum likelihood estimation of long-memory linear processes
- 13 July 2024
- Published by Springer Nature in Statistical Inference for Stochastic Processes
- Vol. 27 (3) , 457-483
Quasi-maximum likelihood estimation of long-memory linear processes
- 11 April 2023
- Published by Wiley in Journal of Time Series Analysis
- Vol. 45 (1) , 103-132
A new estimator for LARCH processes
- 11 April 2023
- Published by Wiley in Journal of Time Series Analysis
- Vol. 45 (1) , 103-132
A new estimator for LARCH processes
- Open Access
A criterion for estimating the largest linear homoscedastic region in Gaussian data
We construct a criterion to identify the largest homoscedastic region in a data set. This can be reduced to a one-sided non-parametric break detection, knowing that up to a certain index the output is governed by a linear homoscedastic model, while after this index it is different ({\it e.g.} a different model, different variables, different volatility, ...). We show the convergence of the estimator of this index, with asymptotic concentration inequalities that can be exponential. Monte Carlo experiments will also confirm its very good numerical performance.
Open AccessA criterion for estimating the largest linear homoscedastic region in Gaussian data
We construct a criterion to identify the largest homoscedastic region in a data set. This can be reduced to a one-sided non-parametric break detection, knowing that up to a certain index the output is governed by a linear homoscedastic model, while after this index it is different ({\it e.g.} a different model, different variables, different volatility, ...). We show the convergence of the estimator of this index, with asymptotic concentration inequalities that can be exponential. Monte Carlo experiments will also confirm its very good numerical performance.
- Open Access
Quasi-Maximum Likelihood Estimation of long-memory linear processes
The purpose of this paper is to study the convergence of the quasi-maximum likelihood (QML) estimator for long memory linear processes. We first establish a correspondence between the long-memory linear process representation and the long-memory AR$(\infty)$ process representation. We then establish the almost sure consistency and asymptotic normality of the QML estimator. Numerical simulations illustrate the theoretical results and confirm the good performance of the estimator.
Open AccessQuasi-Maximum Likelihood Estimation of long-memory linear processes
The purpose of this paper is to study the convergence of the quasi-maximum likelihood (QML) estimator for long memory linear processes. We first establish a correspondence between the long-memory linear process representation and the long-memory AR$(\infty)$ process representation. We then establish the almost sure consistency and asymptotic normality of the QML estimator. Numerical simulations illustrate the theoretical results and confirm the good performance of the estimator.
- 27 December 2022
- Published by Elsevier in Statistics & Probability Letters
- Vol. 195
Laplace’s method and BIC model selection for least absolute value criterion
- 27 December 2022
- Published by Elsevier in Statistics & Probability Letters
- Vol. 195
Laplace’s method and BIC model selection for least absolute value criterion
- Open Access
Contrast estimation of general locally stationary processes using coupling
This paper aims at providing statistical guarantees for a kernel based estimation of time varying parameters driving the dynamic of local stationary processes. We extend the results of Dahlhaus et al. (2018) considering the local stationary version of the infinite memory processes of Doukhan and Wintenberger (2008). The estimators are computed as localized M-estimators of any contrast satisfying appropriate contraction conditions. We prove the uniform consistency and pointwise asymptotic normality of such kernel based estimators. We apply our result to usual contrasts such as least-square, least absolute value, or quasi-maximum likelihood contrasts. Various local-stationary processes as ARMA, AR(infty), GARCH, ARCH(infty), ARMA-GARCH, LARCH(\infty),..., and integer valued processes are also considered. Numerical experiments demonstrate the efficiency of the estimators on both simulated and real data sets.
Open AccessContrast estimation of general locally stationary processes using coupling
This paper aims at providing statistical guarantees for a kernel based estimation of time varying parameters driving the dynamic of local stationary processes. We extend the results of Dahlhaus et al. (2018) considering the local stationary version of the infinite memory processes of Doukhan and Wintenberger (2008). The estimators are computed as localized M-estimators of any contrast satisfying appropriate contraction conditions. We prove the uniform consistency and pointwise asymptotic normality of such kernel based estimators. We apply our result to usual contrasts such as least-square, least absolute value, or quasi-maximum likelihood contrasts. Various local-stationary processes as ARMA, AR(infty), GARCH, ARCH(infty), ARMA-GARCH, LARCH(\infty),..., and integer valued processes are also considered. Numerical experiments demonstrate the efficiency of the estimators on both simulated and real data sets.