			Home About us Contact

Distributed Observations (distributed + observation)

Distribution by Scientific Domains

Mathematics and Statistics	50%

Selected Abstracts

Spatially distributed observations in constraining inundation modelling uncertainties

HYDROLOGICAL PROCESSES, Issue 16 2005
Micha Werner
Abstract The performance of two modelling approaches for predicting floodplain inundation is tested using observed flood extent and 26 distributed floodplain level observations for the 1997 flood event in the town of Usti nad Orlici in the Czech Republic. Although the one-dimensional hydrodynamic model and the integrated one- and two-dimensional model are shown to perform comparably against the flood extent data, the latter shows better performance against the distributed level observations. Comparable performance in predicting the extent of inundation is found to be primarily as a result of the urban reach considered, with flood extent constrained by road and railway embankments. Uncertainty in the elevation model used in both approaches is shown to have little effect on the reliability in predicting flood extent, with a greater impact on the ability in predicting the distributed level observations. These results show that reliability of flood inundation modelling in urban reaches, where flood risk assessment is of more interest than in more rural reaches, can be improved greatly if distributed observations of levels in the floodplain are used in constraining model uncertainties. Copyright © 2005 John Wiley & Sons, Ltd. [source]

Nonparametric two-step regression estimation when regressors and error are dependent

THE CANADIAN JOURNAL OF STATISTICS, Issue 2 2000
Jons Pinkse
Abstract This paper considers estimation of the function g in the model Yt = g(Xt ) + ,t when E(,t|Xt) , 0 with nonzero probability. We assume the existence of an instrumental variable Zt that is independent of ,t, and of an innovation ,t = Xt , E(Xt|Zt). We use a nonparametric regression of Xt on Zt to obtain residuals ,t, which in turn are used to obtain a consistent estimator of g. The estimator was first analyzed by Newey, Powell & Vella (1999) under the assumption that the observations are independent and identically distributed. Here we derive a sample mean-squared-error convergence result for independent identically distributed observations as well as a uniform-convergence result under time-series dependence. Cet article concerne l'estimation de la fonction g dans le modèle Yt = g(Xt) + ,t où E(,t| Xt) , 0 avec probabilité non nulle. Les auteurs supposent l'existence d'une 'variable instrumentale' Zt qui est indépendante de ,t et de l'innovation ,t = Xt , E(Xt|Zt). Les résidus ,t déduits d'une régression non paramétrique de Xt sur Zt permettent d'obtenir une estimation convergente de g. Cette façon de procéder avait déjà été proposée par Newey, Powell & Vella (1999) dans le cas où les observations for-ment un échantillon aléatoire. Les auteurs démontrent ici la convergence de 1'erreur quadratique moyenne expérimentale sous les m,mes conditions et établissent un résultat de convergence uniforme sous des conditions de dépendance sérielle entre les observations. [source]

Regression Calibration in Semiparametric Accelerated Failure Time Models

BIOMETRICS, Issue 2 2010
Menggang Yu
Summary In large cohort studies, it often happens that some covariates are expensive to measure and hence only measured on a validation set. On the other hand, relatively cheap but error-prone measurements of the covariates are available for all subjects. Regression calibration (RC) estimation method (Prentice, 1982,,Biometrika,69, 331,342) is a popular method for analyzing such data and has been applied to the Cox model by Wang et al. (1997,,Biometrics,53, 131,145) under normal measurement error and rare disease assumptions. In this article, we consider the RC estimation method for the semiparametric accelerated failure time model with covariates subject to measurement error. Asymptotic properties of the proposed method are investigated under a two-phase sampling scheme for validation data that are selected via stratified random sampling, resulting in neither independent nor identically distributed observations. We show that the estimates converge to some well-defined parameters. In particular, unbiased estimation is feasible under additive normal measurement error models for normal covariates and under Berkson error models. The proposed method performs well in finite-sample simulation studies. We also apply the proposed method to a depression mortality study. [source]