توابع موثر در مدلهاي خطي ريج با خطا در اندازه گيري تابعي

Ridge estimator has been suggested as an alternative to the maximum likelihood estimator in the presence of collinearity among the elements of unobservable values in functional measurement error models. We derive the influence function of the individual observations on the ridge estimate in this context. We also derive different versions of the sample influence functions and we define the Cookʹs statistic analogues to one given in ordinary linear regression models. Finally, using the influence function we will find the asymptotic distribution of the ridge estimator. Among different results, we show that influence of each observation on the ridge estimator is linear combination of influence on the maximum likelihood estimator.