On Statistical Models for Regression Diagnostics
نویسندگان
چکیده
In regression diagnostics, the case deletion model (CDM) and the mean shift outlier model (MSOM) are commonly used in practice. In this paper we show that the estimates of CDM and MSOM are equal in a wide class of statistical models, which include LSE, MLE, Bayesian estimate and M-estimate in linear and nonlinear regression models; MLE in generalized linear models and exponential family nonlinear models; MLEs of transformation parameters of explanatory variables in a Box-Cox regression models and so on. Furthermore, we study some models, in which, the estimates are not exactly equal but are approximately equal for CDM and MSOM.
منابع مشابه
Categorical Data Regression Diagnostics for Remote Access Servers
Owing to the growing concerns over data confidentiality, many national statistical agencies are considering remote access servers to disseminate data to the public. With remote servers, users submit requests for output from statistical models fit using the collected data, but they are not allowed access to the data. Remote servers also should enable users to check the fit of their models; howev...
متن کاملCook’s distance for ridge estimator in semiparametric regression
The detection of influential observations has attracted a great deal of attention in last few decades. Most of the ideas of determining influential observations are based on single-case diagnostics with ith case deleted. The Cook’s distance are most commonly used among the other single-case diagnostics and successfully applied to various statistical models. In this article, we propose Cook’s di...
متن کاملDynamic linear model diagnostics
In time series analysis using dynamic linear models, retrospective analysis involves the calculation of filtered, or smoothed, distributions for state parameters in the past. We develop and illustrate novel results that are useful in retrospective assessment of the influence of individual observations on such distributions. In particular, new and computationally simple filtering equations are d...
متن کاملInfluence Diagnostics for Correlated Data
Correlated or clustered data are frequently encountered in medical and dental research. Although there has been a large amount of research on analytic methods for clustered data, there has been only a limited amount of work in the area of influence diagnostics for these methods. We present a simple approach to influence diagnostics for clustered data that is practical even for very large data s...
متن کاملTowards a Complete Plasma Diagnostic System
TOWARDS A COMPLETE PLASMA DIAGNOSTIC SYSTEM by Dong Wu Zhao Doctor of Philosophy in Engineering-Electrical Engineering & Computer Sciences University of California at Berkeley Professor Costas J. Spanos, Chair We have set up a plasma diagnostic system with three sources of signals, OES, RF, and machine signals. CF2 OES lines 275 nm and 321 nm are found to be better than any other signals for po...
متن کامل