Dissertations / Theses on the topic 'Correction de biais en ligne'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 22 dissertations / theses for your research on the topic 'Correction de biais en ligne.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Balhane, Saloua. "Improving the dynamical downscaling over Morocco in the context of climate change." Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAX105.
Morocco is one of the most vulnerable regions to climate change. Its climate is characterized by complex interactions between various geographical features, including the Atlantic Ocean, the Mediterranean Sea, the High Atlas Mountains, and the Sahara Desert. Understanding the spatiotemporal variability of climatic patterns in this region is crucial for effective climate change adaptation strategies, natural resource management, and sustainable development planning. Global climate models (GCMs) play a significant role within this context, as they are the only models to take into account all the water and energy reservoirs, including slow-moving reservoirs such as the oceans, which modulate the climate and its evolution. Yet, global climate models are still subject to systematic biases that constrain their performance and have generally coarse resolutions, limiting the assessment of local climate patterns. Regional climate models can improve the representation of certain processes (orographic processes, breezes, etc.). They do, however, have flaws that can significantly alter the credibility of climate change trajectories, as it is impossible to distinguish the impact of systematic biases in the forcing GCMs from the role of better small-scale description.This work explores different ways of overcoming these limitations.In the first part, we evaluate a range of different widely used high-resolution ensembles issued from statistical (NEXGDDP) and dynamical (Euro-CORDEX and bias-adjusted Euro-CORDEX) downscaling while investigating the potential added value that “a posteriori'' bias adjustment may have on the simulation of mean and extreme precipitation and temperature over Morocco.In the second part, we use the LMDZ model, the atmospheric component of the latest version of the IPSL model (IPSLCM6), in a coupled configuration with the ORCHIDEE land-surface model. We designed a refined-grid configuration of the model adapted for regional studies over Morocco that is numerically stable enough for running climate change simulations and allows i) a high resolution over the region and ii) a sufficient resolution on the outside of the zoom area to reproduce large-scale patterns. To deal with the systematic large-scale dynamical biases, a run-time bias correction approach, which consists of bias-correcting the systematic errors in large-scale atmospheric variables using the statistics of a nudged simulation towards climate reanalysis, is used. This method allows for high resolution at a moderate computational cost without compromising the coherence between the global and regional climates. Indeed, preserving this coherence is crucial for Morocco since large-scale circulation patterns play a vital role in shaping regional climate patterns in the region.The evaluation of the present climate (1979–2014) has shown significant improvements after grid refinement, particularly in the mean general circulation. The free refined-grid run compares favorably to precipitation and temperature observations at the local scale. The mean climate is considerably improved after bias correction compared to the uncorrected simulations, and improvements in moisture transport, precipitation and air temperature are observed.For future climate, sea surface temperature (SST) and sea ice concentration (SIC) deduced from four coupled CMIP6 models, forced by greenhouse gases and aerosols corresponding to the Shared Socioeconomic Pathway-8.5 (SSP-8.5) scenario, are used to force the corrected regional configuration of LMDZ6-OR. Twenty-year simulations are produced for a global warming level of 3 Kelvin to assess the response of mean regional climate, precipitation and temperature to changes in SST and SIC
Gelperowic, Céline. "Méthodes de discrimination : comparaison et correction des biais de présélection des échantillons." Paris 9, 2000. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=2000PA090013.
Trinquart, Ludovic. "Impact, détection et correction du biais de publication dans la méta-analyse en réseau." Thesis, Paris 5, 2013. http://www.theses.fr/2013PA05S025/document.
Network meta-analysis (NMA), a generalization of conventional MA, allows for assessing all possible pairwise comparisons between multiple treatments. Reporting bias, a major threat to the validity of MA, has received little attention in the context of NMA. We assessed the impact of reporting bias empirically using data from 74 FDA-registered placebo-controlled trials of 12 antidepressants and their 51 matching publications. We showed how reporting bias biased NMA-based estimates of treatments efficacy and modified ranking. The effect of reporting bias in NMAs may differ from that in classical meta-analyses in that reporting bias affecting only one drug may affect the ranking of all drugs. Then, we extended a test to detect reporting bias in network of trials. It compares the number of expected trials with statistically significant results to the observed number of trials with significant p-values across the network. We showed through simulation studies that the test was fairly powerful after adjustment for size, except when between-trial variance was substantial. Besides, it showed evidence of bias in the network of published antidepressant trials. Finally, we introduced two methods of sensitivity analysis for reporting bias in NMA: a meta-regression model that allows the effect size to depend on its standard error and a selection model that estimates the propensity of trial results being published and in which trials with lower propensity are weighted up in the NMA model. We illustrated their use on the antidepressant datasets. The proposed test and adjustment models borrow strength from all trials across the network, under the assumption that conventional MAs in the network share a common mean bias mechanism
Robin, Yoann. "Transport optimal pour quantifier l'évolution d'un attracteur climatique et corriger ses biais." Thesis, Sorbonne université, 2018. http://www.theses.fr/2018SORUS071/document.
The climate system generates a strange attractor, described by a probability distribution, called the SRB measure (Sinai-Ruelle-Bowen). This measure describes the state and dynamic of the system. The goal of this thesis is first, to quantify the modification of this measure when climate changes. For this, the Wasserstein distance, stemming from the optimal transport theory, allows us determine accurately the differences between probability distributions. Used on a non-autonomous Lorenz toy model, this metric allows us to detect and quantify the alteration due to a forcing similar to anthropogenic forcing. This methodology has been applied to simulation of RCP scenarios from the IPSL model. The results are coherent with different scenarios. Second, the optimal transport gives a theoretical context for stationary bias correction: a bias correction method is equivalent to a joint probability law. A specific joint law is selected with the Wasserstein distance (Optimal Transport Correction method, OTC). This approach allows us extending bias correction methods in any dimension, correcting spatial and inter-variables dependences. An extension in the non-stationary context has been also developed (dynamical OTC method, dOTC). Those two methods have been tested in an idealized case, based on a Lorenz model, and on climate dataset (a regional climate simulation corrected with respect to the SAFRAN reanalysis)
Aligné, Thomas. "Assimilation variationnelle des observations de sondeurs infrarouges hyperspectraux : correction de biais et la détection nuageuse." Toulouse 3, 2007. http://www.theses.fr/2007TOU30038.
Tran, Van-Tinh. "Selection Bias Correction in Supervised Learning with Importance Weight." Thesis, Lyon, 2017. http://www.theses.fr/2017LYSE1118/document.
In the theory of supervised learning, the identical assumption, i.e. the training and test samples are drawn from the same probability distribution, plays a crucial role. Unfortunately, this essential assumption is often violated in the presence of selection bias. Under such condition, the standard supervised learning frameworks may suffer a significant bias. In this thesis, we address the problem of selection bias in supervised learning using the importance weighting method. We first introduce the supervised learning frameworks and discuss the importance of the identical assumption. We then study the importance weighting framework for generative and discriminative learning under a general selection scheme and investigate the potential of Bayesian Network to encode the researcher's a priori assumption about the relationships between the variables, including the selection variable, and to infer the independence and conditional independence relationships that allow selection bias to be corrected.We pay special attention to covariate shift, i.e. a special class of selection bias where the conditional distribution P(y|x) of the training and test data are the same. We propose two methods to improve importance weighting for covariate shift. We first show that the unweighted model is locally less biased than the weighted one on low importance instances, and then propose a method combining the weighted and the unweighted models in order to improve the predictive performance in the target domain. Finally, we investigate the relationship between covariate shift and the missing data problem for data sets with small sample sizes and study a method that uses missing data imputation techniques to correct the covariate shift in simple but realistic scenarios
Robin, Yoann. "Transport optimal pour quantifier l'évolution d'un attracteur climatique et corriger ses biais." Electronic Thesis or Diss., Sorbonne université, 2018. http://www.theses.fr/2018SORUS071.
The climate system generates a strange attractor, described by a probability distribution, called the SRB measure (Sinai-Ruelle-Bowen). This measure describes the state and dynamic of the system. The goal of this thesis is first, to quantify the modification of this measure when climate changes. For this, the Wasserstein distance, stemming from the optimal transport theory, allows us determine accurately the differences between probability distributions. Used on a non-autonomous Lorenz toy model, this metric allows us to detect and quantify the alteration due to a forcing similar to anthropogenic forcing. This methodology has been applied to simulation of RCP scenarios from the IPSL model. The results are coherent with different scenarios. Second, the optimal transport gives a theoretical context for stationary bias correction: a bias correction method is equivalent to a joint probability law. A specific joint law is selected with the Wasserstein distance (Optimal Transport Correction method, OTC). This approach allows us extending bias correction methods in any dimension, correcting spatial and inter-variables dependences. An extension in the non-stationary context has been also developed (dynamical OTC method, dOTC). Those two methods have been tested in an idealized case, based on a Lorenz model, and on climate dataset (a regional climate simulation corrected with respect to the SAFRAN reanalysis)
Moussaoui, Mohamed. "Optimisation de la correction de biais dans le récepteur PIC multi-étages pour le système CDMA." Valenciennes, 2005. http://ged.univ-valenciennes.fr/nuxeo/site/esupversions/19958d9e-d214-4166-b25a-481ff9aca6ff.
The complexity of Verdu's optimal receiver for CDMA increases exponentially with the number of users, loading to an unrealistic implementation. In this thesis, we analyze the sub-optimal interference cancellation receiver with multi-stages parallel structure. The parallel nature of the algorithm can be easily exploited in a multiprocessing environment. This makes it extremely attracting for UMTS-TDD. The use of a matched-filter estimator, results in a bias in the estimated amplitude at the second stage output, particularly for heavy system loads. This bias degrades the system performances in term of bit error rate (BER). We propose an original low complexity approach for reducing the bias, in which we don't attenuate the estimated multiple access interference (MAI), as in the partial cancellation method, but we amplify the amplitude of the received signal for each user by a scalar which we call the amplification factor (AF). An analytical study in the synchronous case and by simulation in the asynchronous case was led to determine an optimal value of this amplification factor, in order to minimize the BER. We indicate the performances obtained in a perfect power control case and for uncompensated near-far effects. Then, we propose a structure incorporating this amplification factor in the multi-stage case, and we compare its complexity with that for the partial cancellation solution. Another aspect considered in this thesis is the impact of the decision functions on the performance of the PIC receiver. We thus considered the `null zone' and ` clipping' functions and compared their sensitivity to decision threshold optimization errors
Li, Meng. "Développement de l'expression orale du français chez les apprenants chinois par le biais de l'échange écrit instantané en ligne." Thesis, Limoges, 2020. http://aurore.unilim.fr/theses/nxfile/default/e30e642f-7e77-4c68-a010-2f57edcde34d/blobholder:0/2020LIMO0058.pdf.
This research concerns teaching French as a foreign language. It focuses exclusively on the development of French oral expression skills by way of synchronous text-based online exchange. The research began when the authors noticed that oral expression was a big handicap for Chinese students and that teaching practices in the classroom were insufficient to remedy it. Against this background, we referred to Information and Communication Technologies for Education (ICTE) and asked the question: could ICTE help create a teaching / learning device in order to improve oral activities, complement spoken French teaching, and improve significantly Chinese students’ oral expression skills? To answer this question, this paper focuses on a key pedagogical issue, by reference to the hypothesis that, as an auxiliary method, synchronous text-based online exchange could assist Chinese students majoring in French to improve their speaking skills. Focusing on both key learning theories and teaching information and communication technologies, examining this hypothesis permitted us to analyse the close relationship between written and oral expression in order to improve Chinese students’ oral expression skills. Results of our teaching experiment further confirmed the hypothesis, while also opening up new avenues of reflection on this issue and suggesting future research
Fourati, Mariem. "Étude de la confusion résiduelle et erreur de mesure dans les modèles de régression." Mémoire, Université de Sherbrooke, 2015. http://hdl.handle.net/11143/6983.
Leang, Isabelle. "Fusion en ligne d'algorithmes de suivi visuel d'objet." Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066486/document.
Visual object tracking is an elementary function of computer vision that has been the subject of numerous studies. Drift over time is one of the most critical phenomena to master because it leads to the permanent loss of the target being tracked. Despite the numerous approaches proposed in the literature to counter this phenomenon, none outperforms another in terms of robustness to the various sources of visual perturbations: variation of illumination, occlusion, sudden movement of camera, change of aspect. The objective of this thesis is to exploit the complementarity of a set of tracking algorithms by developing on-line fusion strategies capable of combining them generically. The proposed fusion chain consists of selecting the trackers from indicators of good functioning, combining their outputs and correcting them. On-line drift prediction was studied as a key element of the selection mechanism. Several methods are proposed for each step of the chain, giving rise to 46 possible fusion configurations. Evaluated on 3 databases, the study highlighted several key findings: effective selection greatly improves robustness; The correction improves the robustness but is sensitive to bad selection, making updating preferable to reinitialization; It is more advantageous to combine a small number of complementary trackers with homogeneous performances than a large number; The robustness of fusion of a small number of trackers is correlated to the incompleteness measure, which makes it possible to select the appropriate combination of trackers to a given application context
Yameogo, Nadège-Désirée. "Analyse de la demande résidentielle d'électricité à partir d'enquêtes indépendantes : correction de biais de sélection et d'endogénéité dans un contexte de classes latentes." Thesis, Université Laval, 2008. http://www.theses.ulaval.ca/2008/25076/25076.pdf.
Yaméogo, Nadège-Désirée. "Analyse de la demande résidentielle d'électricité à partir d'enquêtes indépendantes : correction de biais de sélection et d'endogénéité dans un contexte de classes latentes." Doctoral thesis, Université Laval, 2008. http://hdl.handle.net/20.500.11794/19745.
Moua, Yi. "Correction de l’effet du biais d’échantillonnage dans la modélisation de la qualité des habitats écologiques : application au principal vecteur du paludisme en Guyane française." Thesis, Guyane, 2017. http://www.theses.fr/2017YANE0002/document.
Species distribution models are identified as relevant to map and characterize the habitat quality of Anopheles genusmosquitoes, transmitting malaria, and thus to both participate in the estimation of the transmission risk of this disease and inthe definition of targeted vector control actions. The malaria transmission depends on the presence and distribution of thevectors, which are themselves dependent on the environmental conditions that define the quality of the ecological habitats of the Anopheles. However, in some areas, Anopheles collection data remain scarce, making it difficult to model these habitats. In addition, the collection of these data is very often subjected to significant sampling biases, due, in particular, to unequal accessibility to the entire study area. This thesis provides a solution to the mapping of malaria vectors, considering two very few studied aspects in modeling: the low number of available presence sites and the existence of a sampling bias. An original method for correcting the effect of the sampling bias is proposed and then applied to presence data of Anopheles darlingi species - the main vector of malaria in South America - in French Guiana. Then, a distribution model of An. darlingi was built to obtain a map of habitat quality consistent with entomologists’ knowledge and providing high prediction performances. The proposed correction method was then compared to existing methods in an application context characterized by the scarcity of the species occurence data and the presence of a sampling bias. The results show that the developed method is adapted to cases where the number of sites of presence is low. This thesis contributes, on the one hand, to fill theoretical and applicability lacuna of current methods intended to correct the effect of the sampling bias and, on the other hand, to supplement the knowledge on both the spatial distribution and the bio-ecology of the main malaria vector in French Guiana
Py, Julien. "Modélisation et développement d'un système d'analyse en ligne des transuraniens par spectrométrie de fluorescence X raies L." Thesis, Besançon, 2014. http://www.theses.fr/2014BESA2047/document.
This thesis deals with the development of a new compact, accurate, fast, without cooling liquid, fluorescence L X-ray spectrometer, with the aim to determinate online transuranic elements (uranium, plutonium and americium) in nuclear materials reprocessing. The objective was to define the configuration and the characteristics of this spectrometer and the method to quantify transuranic elements between 0.1 g/L and 20 g/L. To minimize as law as possible the manipulation of these elements, we have used an original approach, namely Monte-Carlo simulations and none radioactive surrogate elements. The study of these solutions allowed to eliminate the specific effects of transuranic elements decay (internal conversion) and to optimize the spectrometer. Monte-Carlo simulations with the PENELOPE code were used for two reasons. Firstly, we have developed an analytical method to correct the matrix effects. Then, we have selected three systems to produce a quasi-monochromatic X-rays beam from the X-ray generator, to optimize the intensity of the L X-ray fluorescence spectra, and to measure the Compton scatter peak. These systems were then tested with an adjustable spectrometer in order to select and optimize the best configuration. We have shown that, after peaks and left tails separation from spectra with COLEGRAM software, the spectrometer can be used to analyze solutions with various concentrations of thallium and bismuth. Several solutions of uranium or plutonium were analyzed to determine the effects of interferences from gamma rays and internal conversion with the fluorescence L X-rays. These effects have been corrected by subtracting a passive spectrum to an active one
Vaittinada, ayar Pradeebane. "Intercomparaison et développement de modèles statistiques pour la régionalisation du climat." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLV010/document.
The study of climate variability is vital in order to understand and anticipate the consequences of future climate changes. Large data sets generated by general circulation models (GCMs) are currently available and enable us to conduct studies in that direction. However, these models resolve only partially the interactions between climate and human activities, namely du to their coarse resolution. Nowadays there is a large variety of models coping with this issue and aiming at generating climate variables at local scale from large-scale variables : the downscaling models.The aim of this thesis is to increase the knowledge about statistical downscaling models (SDMs) wherein there is many approaches. The work conducted here pursues four main goals : (i) to discriminate statistical (and dynamical) downscaling models, (ii) to study the influences of GCMs biases on the SDMs through a bias correction scheme, (iii) to develop a statistical downscaling model accounting for climate spatial and temporal non-stationarity in a spatial modelling context and finally, (iv) to define seasons thanks to a weather typing modelling.The intercomparison of downscaling models led to set up a model selection methodology according to the end-users needs. The study of the biases of the GCMs reveals the impacts of those biases on the SDMs simulations and the positive contributions of the bias correction procedure. The different steps of the spatial SDM development bring some interesting and encouraging results. The seasons defined by the weather regimes are relevant for seasonal analyses and modelling.All those works conducted in a “Statistical Climatologie” framework lead to many relevant perspectives, not only in terms of methodology or knowlegde about local-scale climate, but also in terms of use by the society
Mazet, Vincent. "Développement de méthodes de traitement de signaux spectroscopiques : estimation de la ligne de base et du spectre de raies." Phd thesis, Université Henri Poincaré - Nancy I, 2005. http://tel.archives-ouvertes.fr/tel-00011477.
Dans un premier temps est proposée une méthode déterministe qui permet d'estimer la ligne de base des spectres par le polynôme qui minimise une fonction-coût non quadratique (fonction de Huber ou parabole tronquée). En particulier, les versions asymétriques sont particulièrement bien adaptées pour les spectres dont les raies sont positives. Pour la minimisation, on utilise l'algorithme de minimisation semi-quadratique LEGEND.
Dans un deuxième temps, on souhaite estimer le spectre de raies : l'approche bayésienne couplée aux techniques MCMC fournit un cadre d'étude très efficace. Une première approche formalise le problème en tant que déconvolution impulsionnelle myope non supervisée. En particulier, le signal impulsionnel est modélisé par un processus Bernoulli-gaussien à support positif ; un algorithme d'acceptation-rejet mixte permet la simulation de lois normales tronquées. Une alternative intéressante à cette approche est de considérer le problème comme une décomposition en motifs élémentaires. Un modèle original est alors introduit ; il a l'intérêt de conserver l'ordre du système fixe. Le problème de permutation d'indices est également étudié et un algorithme de ré-indexage est proposé.
Les algorithmes sont validés sur des spectres simulés puis sur des spectres infrarouge et Raman réels.
Vaittinada, ayar Pradeebane. "Intercomparaison et développement de modèles statistiques pour la régionalisation du climat." Electronic Thesis or Diss., Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLV010.
The study of climate variability is vital in order to understand and anticipate the consequences of future climate changes. Large data sets generated by general circulation models (GCMs) are currently available and enable us to conduct studies in that direction. However, these models resolve only partially the interactions between climate and human activities, namely du to their coarse resolution. Nowadays there is a large variety of models coping with this issue and aiming at generating climate variables at local scale from large-scale variables : the downscaling models.The aim of this thesis is to increase the knowledge about statistical downscaling models (SDMs) wherein there is many approaches. The work conducted here pursues four main goals : (i) to discriminate statistical (and dynamical) downscaling models, (ii) to study the influences of GCMs biases on the SDMs through a bias correction scheme, (iii) to develop a statistical downscaling model accounting for climate spatial and temporal non-stationarity in a spatial modelling context and finally, (iv) to define seasons thanks to a weather typing modelling.The intercomparison of downscaling models led to set up a model selection methodology according to the end-users needs. The study of the biases of the GCMs reveals the impacts of those biases on the SDMs simulations and the positive contributions of the bias correction procedure. The different steps of the spatial SDM development bring some interesting and encouraging results. The seasons defined by the weather regimes are relevant for seasonal analyses and modelling.All those works conducted in a “Statistical Climatologie” framework lead to many relevant perspectives, not only in terms of methodology or knowlegde about local-scale climate, but also in terms of use by the society
Morillot, Olivier. "Reconnaissance de textes manuscrits par modèles de Markov cachés et réseaux de neurones récurrents : application à l'écriture latine et arabe." Electronic Thesis or Diss., Paris, ENST, 2014. http://www.theses.fr/2014ENST0002.
Handwriting recognition is an essential component of document analysis. One of the popular trends is to go from isolated word to word sequence recognition. Our work aims to propose a text-line recognition system without explicit word segmentation. In order to build an efficient model, we intervene at different levels of the recognition system. First of all, we introduce two new preprocessing techniques : a cleaning and a local baseline correction for text-lines. Then, a language model is built and optimized for handwritten mails. Afterwards, we propose two state-of-the-art recognition systems based on contextual HMMs (Hidden Markov Models) and recurrent neural networks BLSTM (Bi-directional Long Short-Term Memory). We optimize our systems in order to give a comparison of those two approaches. Our systems are evaluated on arabic and latin cursive handwritings and have been submitted to two international handwriting recognition competitions. At last, we introduce a strategy for some out-of-vocabulary character strings recognition, as a prospect of future work
Ringard, Justine. "Estimation des précipitations sur le plateau des Guyanes par l'apport de la télédétection satellite." Thesis, Guyane, 2017. http://www.theses.fr/2017YANE0010/document.
The Guiana Shield is a region that is characterized by 90% of a primary rainforest and about 20% of the world’s freshwater reserves. This natural territory, with its vast hydrographic network, shows annual rainfall intensities up to 4000 mm/year; making this plateau one of the most watered regions in the world. In addition, tropical rainfall is characterized by significant spatial and temporal variability. In addition to climate-related aspects, the impact of rainfall in this region of the world is significant in terms of energy supply (hydroelectric dams). It is therefore important to develop tools to estimate quantitatively and qualitatively and at high spatial and temporal resolution the precipitation in this area. However, this vast geographical area is characterized by a network of poorly developed and heterogeneous rain gauges, which results in a lack of knowledge of the precise spatio-temporal distribution of precipitation and their dynamics.The work carried out in this thesis aims to improve the knowledge of precipitation on the Guiana Shield by using Satellite Precipitation Product (SPP) data that offer better spatial and temporal resolution in this area than the in situ measurements, at the cost of poor quality in terms of precision.This thesis is divided into 3 parts. The first part compares the performance of four products of satellite estimates on the study area and attempts to answer the question : what is the quality of these products in the Northern Amazon and French Guiana in spatial and time dimensions ? The second part proposes a new SPP bias correction technique that proceeds in three steps: i) using rain gauges measurements to decompose the studied area into hydro climatic areas ii) parameterizing a bias correction method called quantile mapping on each of these areas iii) apply the correction method to the satellite data for each hydro-climatic area. We then try to answer the following question : does the parameterization of the quantile mapping method on different hydro-climatic areas make it possible to correct the precipitation satellite data on the study area ? After showing the interest of taking into account the different rainfall regimes to implement the QM correction method on SPP data, the third part analyzes the impact of the temporal resolution of the precipitation data used on the quality of the correction and the spatial extent of potentially correctable SPP data (SPP data on which the correction method can be applied effectively). In summary, the objective of this section is to evaluate the ability of our method to correct on a large spatial scale the bias of the TRMM-TMPA 3B42V7 data in order to make the exploitation of this product relevant for different hydrological applications.This work made it possible to correct the daily satellite series with high spatial and temporal resolution on the Guiana Shield using a new approach that uses the definition of hydro-climatic areas. The positive results in terms of reduction of the bias and the RMSE obtained, thanks to this new approach, makes possible the generalization of this new method in sparselygauged areas
Richard, Michael. "Évaluation et validation de prévisions en loi." Thesis, Orléans, 2019. http://www.theses.fr/2019ORLE0501.
In this thesis, we study the evaluation and validation of predictive densities. In a first part, we are interested in the contribution of machine learning in the field of quantile and densityforecasting. We use some machine learning algorithms in quantile forecasting framework with real data, inorder to highlight the efficiency of particular method varying with nature of the data.In a second part, we expose some validation tests of predictive densities present in the literature. Asillustration, we use two of the mentionned tests on real data concerned about stock indexes log-returns.In the third part, we address the calibration constraint of probability forecasting. We propose a generic methodfor recalibration, which allows us to enforce this constraint. Thus, it permits to simplify the choice betweensome density forecasts. It remains to be known the impact on forecast quality, measured by predictivedistributions sharpness, or specific scores. We show that the impact on the Continuous Ranked ProbabilityScore (CRPS) is weak under some hypotheses and that it is positive under more restrictive ones. We use ourmethod on weather and electricity price ensemble forecasts.Keywords : Density forecasting, quantile forecasting, machine learning, validity tests, calibration, bias correction,PIT series , Pinball-Loss, CRPS
Mackrous, Isabelle. "Rôle de la vision pour le contrôle de la dynamique du mouvement lors d'un geste de pointage manuel chez l'adulte ainsi que chez l'enfant." Thèse, 2009. http://hdl.handle.net/1866/6473.