کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
416463 681370 2012 11 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Computing highly accurate or exact PP-values using importance sampling
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر نظریه محاسباتی و ریاضیات
پیش نمایش صفحه اول مقاله
Computing highly accurate or exact PP-values using importance sampling
چکیده انگلیسی

Especially for discrete data, standard first order PP-values can suffer from poor accuracy, even for quite large sample sizes. Moreover, different test statistics can give practically different results. There are several approaches to computing PP-values which do not suffer these defects, such as parametric bootstrap PP-values or the partially maximised PP-values of Berger and Boos (1994).Both these methods require computing the exact tail probability of the approximate PP-value as a function of the nuisance parameter/s, known as the significance profile. For most practical problems, this is not computationally feasible. I develop an importance sampling approach to this problem. A major advantage is that significance can be simultaneously estimated at a grid of nuisance parameter values, without the need for smoothing away the simulation noise. The theory is fully developed for generalised linear models. The importance distribution is selected from the same generalised linear model family but with parameters biased towards an optimal point on the boundary of the tail-set. For logistic regression at least, standard guidelines for selecting the importance distribution can fail quite badly and a conceptually simple alternative algorithm for selecting these parameters is developed. This may have application to importance sampling more generally.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Computational Statistics & Data Analysis - Volume 56, Issue 6, June 2012, Pages 1784–1794
نویسندگان
,