| Titre : | Contribution to the nonparametric estimation in relation with the truncated and dependent data |
| Auteurs : | Menasria Nedjeh, Auteur ; Bouazza Imane, Directeur de thèse |
| Type de document : | texte manuscrit |
| Editeur : | [S.l.] : [S.l.] : [S.l.] : Université Saïda – Dr. Tahar Moulay – Faculté des Mathématiques, de l’Informatique et de Télécommunications, 2024/2025 |
| Format : | 64ص |
| Accompagnement : | CD |
| Langues: | Anglais |
| Index. décimale : | BUC-M 003690 |
| Catégories : |
Master Mathématiques: Specialty: Stochastic Analysis, Process Statistics and Applications (SASPA |
| Résumé : |
In conducting this work, we acknowledge that the asymptotic properties of non-
parametric conditional models associated with the recursive kernel approach have con- sistently been a focus of researchers; consequently, they occasionally presume that the sample under investigation comprises functional α mixing observations. In order to achieve this work, we started with Chapter 1 to review the key principles and concepts related to statistics such as recursive method, dependent and incomplete data, as well as other features required for the research, namely truncated data. 4.2 Conclusion 59 The focus of Chapter 2 and 3 was then mostly on highlighting the challenges that arise when we attempt to deal with issues in particular truncated and dependent data, employing an estimator of the conditional distribution function and a quantile that is based on the recursive technique. Firstly, we use the recursive double-kernel approach to estimate the conditional distribution function for complete sample of random variables when the variable of interest is real. Secondly, by using the same approach we tried to estimate the conditional distribution function for an incomplete sample (truncated data) of variables when the variable of interest is a vector. In the meantime of the both cases, we performed the almost sure convergence with the estimators’ rates. Finally, Chapter 4 came to verify the theoretical findings for an infinite sample size and various truncation rates, we attempted to simulate at the end of this study in the purpose to examining the numerical behavior of the estimators. |
Exemplaires
| Code-barres | Cote | Support | Localisation | Section | Disponibilité |
|---|---|---|---|---|---|
| aucun exemplaire |
Documents numériques (1)
BUC-M 003690 Adobe Acrobat PDF |
