No abstract available

No abstract available

[en] In this paper the testing of normality for un unconditionally heterosedastic macroeconomic time series is studied. It is underlined that the classic al Jarque-Bera test (JB hereafter) for normality is inadequate in our framework. On the other hand it is found that the approach which consists in correcting the heteroscedasticity by kernel smoothing for testing normality is justified asymptotically. Nevertheless it appears from Monte Carlo experiments that such a methodology can noticeably suffer from size distortion for samples that are typical for macroeconomic variables. As a consequence a parametric bootstrap methodology for correcting the problem is proposed. The innovations distribution of a set of inflation measures for the U.S., Korea and Australia are analyzed.

[en] The orbital motion of an artificial satellite or space debris object is perturbed by a variety, and sometimes not well-modeled, external forces [1, 2]. The hybrid methodology can be used to predict these unmodeled effects or the uncertainty associated with this process. In this work, a Hybrid Orbit Propagator based on SGP4 [3–7] and a state space formulation of the exponential smoothing method as the forecasting technique is developed. The error terms of the forecasting technique are considered Gaussian noise what allows us to use the maximum likelihood method to estimate the parameters of the exponential smoothing model, as well as computing the point forecast and the reliable predictive intervals. Finally, this Hybrid Orbit Propagator is applied to data from a satellite of the Galileo constellation. This Propagator improves the accuracy of the classical SGP4 and it is particularly good for short forecast horizons.

[en] A ''dynamo theory'' technique appears to successfully predict decadal time scale solar activity variations. The technique was developed a decade ago, following some puzzling correlations involved with geomagnetic ''precursors'' of solar activity. Based upon this, a dynamo theory method was developed to predict solar activity. The method was used successfully in solar cycle. We now see sunspot numbers increasing at an alarming rate and realize 1) that a large cycle is developing and 2) that the cycle may even surpass the largest cycle-19. We use a ''Sporer Butterfly'' method to show that the cycle can now be expected to peak in the latter half of 1989, consistent with an amplitude comparable to our value predicted near the last solar minimum

[en] In this paper we propose a methodology to build a model for predicting future outbreaks of Methicilin-resistant Staphylococcus aereus (MRSA). Infection incidence forecasting is approached as a feature selection based time series forecasting problem using multivariate time series composed of incidence of Staphylococcus Aereus and MRSA infections, influenza incidence and total days of therapy of both of Levoflaxin and Oseltamivir antimicrobials.

[en] Flash floods frequently hit Southern France and cause heavy damages and fatalities. To better protect persons and goods, official flood forecasting services in France need accurate information and efficient models to optimize their decision and policy. Since heavy rainfalls that cause such floods are very heterogeneous, it becomes a serious challenge for forecasters. Such phenomena are typically nonlinear and more complex than classical floods events. That problem leads to consider complementary alternatives to enhance the management of such situations. For decades, artificial neural networks have been very efficient to model nonlinear phenomena, particularly rainfall-discharge relations in various types of basins. They are applied in this study with two main goals: first modelling flash floods on the Gardon de Mialet basin; second, extract internal information from the model by using the Knowledge eXtraction method to provide new ways to improve models. The first analysis shows that the kind of nonlinear predictor influences strongly the representation of information: e.g. the main influent variable (rainfall) is more important in the recurrent and static models than in the feed-forward one. For understanding flash floods genesis, recurrent and static models appear thus as better candidates, even if their results are better.

[en] This paper considers modelling of a non-stationary integervalued autoregressive moving average of order 1 (INARMA(1,1)) model by assuming that the innovation follows a Poisson and negative binomial distribution. Two simulation experiments are also conducted to assess the performance of conditional maximum likelihood (CML) and generalized quasi-likelihood (GQL) estimation methods.

[en] Nowadays, most technical devices generate and store data in the form of time series. They are widely used in climatology, where e.g. water levels are collected for dike construction, in agriculture or the energy sector, for water management, and many other industries. Such time series help farmers, for example, to calculate the risk of frost in April, or an energy company to estimate the photovoltaic production, which - among other things - depends on the temperature.

[en] This paper considers the approximate factor model for high-dimensional time series with additive outliers. We propose a robustification procedure of the information criteria proposed by [1]. The robust estimator of the number of factors is obtained by replacing the standard covariance matrix with M-covariance matrix. Simulations are carried out under the scenarios of multivariate time series with and without additive outliers to assess the impact of additive outliers on the standard information criteria and to analyze the finite sample size performance of the proposed robust estimator of the number of factors.