Some Faults of the Bayes Factor in Nonparametric Model Selection

We compare different Bayesian strategies for testing a parametric model versus a nonparametric alternative on the ground of their ability to solve the inconsistency problems arising when using the Bayes factor under certain conditions. A preliminary critical discussion of such an inconsistency is provided.     

LOCAL POLYNOMIAL QUASI-LIKELIHOOD REGRESSION ON RANDOM FIELDS

In this paper, local quasi‐likelihood regression is considered for stationary random fields of dependent variables. In the case of independent data, local polynomial quasi‐likelihood regression is known to have several appealing features such as minimax efficiency, design adaptivity and good boundary behaviour. These properties are shown to carry over to the case of random fields. The asymptotic normality of the regression estimator is established and explicit formulae for its asymptotic bias and variance are derived for strongly mixing stationary random fields. The extension to multi‐dimensional covariates is also provided in full generality. Moreover, evaluation of the finite sample performance is made through a simulation study.     

Maximum likelihood estimation of linear continuous time long memory processes with discrete time data

Summary.  We develop a new class of time continuous autoregressive fractionally integrated moving average (CARFIMA) models which are useful for modelling regularly spaced and irregu-larly spaced discrete time long memory data. We derive the autocovariance function of a stationary CARFIMA model and study maximum likelihood estimation of a regression model with CARFIMA errors, based on discrete time data and via the innovations algorithm. It is shown that the maximum likelihood estimator is asymptotically normal, and its finite sample properties are studied through simulation. The efficacy of the approach proposed is demonstrated with a data set from an environmental study.     

Local Likelihood Estimation in Generalized Additive Models

ABSTRACT.  Generalized additive models are a popular class of multivariate non-parametric regression models, due in large part to the ease of use of the local scoring estimation algorithm. However, the theoretical properties of the local scoring estimator are poorly understood. In this article, we propose a local likelihood estimator for generalized additive models that is closely related to the local scoring estimator fitted by local polynomial regression. We derive the statistical properties of the estimator and show that it achieves the same asymptotic convergence rate as a one-dimensional local polynomial regression estimator. We also propose a wild bootstrap estimator for calculating point-wise confidence intervals for the additive component functions. The practical behaviour of the proposed estimator is illustrated through a simulation experiment.     

Modelling forces of infection by using monotone local polynomials

Summary. On the basis of serological data from prevalence studies of rubella, mumps and hepatitis A, the paper describes a flexible local maximum likelihood method for the estimation of the rate at which susceptible individuals acquire infection at different ages. In contrast with parametric models that have been used before in the literature, the local polynomial likelihood method allows this age-dependent force of infection to be modelled without making any assumptions about the parametric structure. Moreover, this method allows for simultaneous nonparametric estimation of age-specific incidence and prevalence. Unconstrained models may lead to negative estimates for the force of infection at certain ages. To overcome this problem and to guarantee maximal flexibility, the local smoother can be constrained to be monotone. It turns out that different parametric and nonparametric estimates of the force of infection can exhibit considerably different qualitative features like location and the number of maxima, emphasizing the importance of a well-chosen flexible statistical model.     

Bayesian and maximin optimal designs for heteroscedastic regression models

The authors consider the problem of constructing standardized maximin D‐optimal designs for weighted polynomial regression models. In particular they show that by following the approach to the construction of maximin designs introduced recently by Dette, Haines & Imhof (2003), such designs can be obtained as weak limits of the corresponding Bayesian _q‐optimal designs. They further demonstrate that the results are more broadly applicable to certain families of nonlinear models. The authors examine two specific weighted polynomial models in some detail and illustrate their results by means of a weighted quadratic regression model and the Bleasdale–Nelder model. They also present a capstone example involving a generalized exponential growth model.     

Fractional Factorial Designs for the Detection of Interactions between Design and Noise Factors

In industrial experiments on both design (control) factors and noise factors aimed at improving the quality of manufactured products, designs are needed which afford independent estimation of all design×noise interactions in as few runs as possible, while allowing aliasing between those factorial effects of less interest. An algorithm for generating orthogonal fractional factorial designs of this type is described for factors at two levels. The generated designs are appropriate for experimenting on individual factors or for experimentation involving group screening of factors.     

Automatic repeated-loess decomposition of data consisting of sums of oscillatory curves

Repeated loess is a nonparametric procedure that uses progressive smoothing and differencing to decompose data consisting of sums of curves. Smoothing is by locally weighted polynomial regression. Here the procedure was developed so that the decomposition into components was controlled automatically by the number of maxima in each component. The level of smoothing of each component was chosen to maximize the estimated probability of the observed number of maxima. No assumptions were made about the periodicity of components and only very weak assumptions about their shapes. The automatic procedure was applied to simulated data and to experimental data on human visual sensitivity to line orientation.An erratum to this article can be found at     

The role of the drift in I(2) systems

Summary This paper discussed the role of the drift in vector autoregressive processes allowing for integrated components up to order 2. It is shown how the drift can generate linear and quadratic deterministic trends. A two-stage statistical analysis of the system in the presence of quadratic trends is proposed. The analysis of the system allows to define a consistent sequence of tests on the numbers of common components integrated of a given order, called the integration indices of the system. The relevant asymptotic distributions are non-standard, belong to the Limiting Gaussian Functional family and are tabulated by simulation. The proposed procedure can also be consistently combined with other procedures proposed by the author for the cases of a linear trend and of no deterministic trends in the system. Invited paper at the Conference held in Bologna, Italy, 27–28 May 1993, on “Statistical Tests: Methodology and Econometric Applications”. Partial financial support is acknowledged from Italian MURST grants 40% and 60%.     

Steffensen不等式的推广

本文证明了Ｓｔｅｆｆｅｎｓｅｎ不等式的一个新的推广形式．