![]() | يوجد فقط 14 صفحة متاحة للعرض العام |
المستخلص In this dissertation, the researcher is interested in examining the linear and non-linear calibrationproblem. While the problem of linear calibration has four cases each of which has different assumptions, the researcher has chosen to study the problem of nonlinear calibration with a special focus on the quadratic model as a special case; which could be easily generalized to cases of higher order. In particular, the researcher proposes a novel technique based on the idea of combining both the conventional resampling methods with Gibbs sampling technique in order to estimate and infer the unknownx_0. For this purpose, the researcher has used three methods of resampling. These methods are bootstrap, smoothed bootstrap and Jackknife-after-bootstrap. The main idea here is to improve the statistical properties of the estimated x_0such as variance and confidence interval, compared with the conventional methods of linear and nonlinear calibration. |