Fixed effects ANOVA: an extension to samples with random size

In many relevant situations, such as in medical research, sample sizes may not be previously known. The aim of this paper is to extend one and more than one-way analysis of variance to those situations and show how to compute correct critical values. The interest of this approach lies in avoiding false rejections obtained when using the classical fixed size F-tests. Sample sizes are assumed as random and we then proceed with the application of this approach to a database on cancer.     

OPTIMAL DESIGN FOR EARLY‐GENERATION PLANT‐BREEDING TRIALS WITH UNREPLICATED OR PARTIALLY REPLICATED TEST LINES

Some subtle difficulties in optimal design are highlighted by the example of unreplicated field trials laid out on plots with spatial errors defined by uniformity trials. There is a dual problem of the arrangement of control plots and maximizing the number of test‐line entries. A simulation study is conducted by randomizing the allocation of genotypes to the plots of four uniformity trials in accordance with the rules defining a number of competing designs. Results are summarized in terms of the ‘SE ratio’, which reflects the improvement in precision of a given design relative to a completely random design on the same plots. The definition of the SE ratio overcomes problems induced by differential shrinkage and consequent precision of test and control lines. A general result applying to all designs shows a curvilinear improvement in SE ratio with increasing error degrees of freedom of the design. The actual arrangement of check plots is of less importance than their increasing number, which contributes to increasing error degrees of freedom. Overall measures, including expected genetic gain, are used to illustrate the choice of a balance between the total number of test‐line entries and the error degrees of freedom.     

Autoregressive model selection based on a prediction perspective

The autoregressive (AR) model is a popular method for fitting and prediction in analyzing time-dependent data, where selecting an accurate model among considered orders is a crucial issue. Two commonly used selection criteria are the Akaike information criterion and the Bayesian information criterion. However, the two criteria are known to suffer potential problems regarding overfit and underfit, respectively. Therefore, using them would perform well in some situations, but poorly in others. In this paper, we propose a new criterion in terms of the prediction perspective based on the concept of generalized degrees of freedom for AR model selection. We derive an approximately unbiased estimator of mean-squared prediction errors based on a data perturbation technique for selecting the order parameter, where the estimation uncertainty involved in a modeling procedure is considered. Some numerical experiments are performed to illustrate the superiority of the proposed method over some commonly used order selection criteria. Finally, the methodology is applied to a real data example to predict the weekly rate of return on the stock price of Taiwan Semiconductor Manufacturing Company and the results indicate that the proposed method is satisfactory.     

Testable hypotheses in singular fixed linear models

Hypotheses that restrict nonestimable parameters in singular (or overparameterized) fixed linear models are considered nontes table by most aothors and are not allowed by most computer packages. In this article, a different approach is taken and hypotheses are classified as completely testable, partially testable, or nontestable on the basis of the number of degrees of freedom associated with them. The convenience of this approach is illustrated with examples and by developing a related general theory of equivalent hypotheses, reparameterizations, and restrictions. A method of transforming partially testable hypotheses into equivalent completely testable hypotheses is described.     

Nonparametric and non-linear models and data mining in time series: a case-study on the Canadian lynx data

Nonparametric regression methods are used as exploratory tools for formulating, identifying and estimating non-linear models for the Canadian lynx data, which have attained bench-mark status in the time series literature since the work of Moran in 1953. To avoid the curse of dimensionality in the nonparametric analysis of this short series with 114 observations, we confine attention to the restricted class of additive and projection pursuit regression (PPR) models and rely on the estimated prediction error variance to compare the predictive performance of various (non-)linear models. A PPR model is found to have the smallest (in-sample) estimated prediction error variance of all the models fitted to these data in the literature. We use a data perturbation procedure to assess and adjust for the effect of data mining on the estimated prediction error variances; this renders most models fitted to the lynx data comparable and nearly equivalent. However, on the basis of the mean-squared error of out-of-sample prediction error, the semiparametric model X_t =1.08+1.37 X_{t −1}+ f ( X_{t −2})+ e_t and Tong's self-exciting threshold autoregression model perform much better than the PPR and other models known for the lynx data.     

A note on the power of the f-test for testing linear hypothesis with elliptic t errors

The power of the classical .F-test for testing the regression coefficient of a general linear model with elliptic t error variable depends on the degrees of freedom of the t- distribution. In this note it is shown that the power of the F-test based on t-distribution is greater than the normal based test at smaller level of significance.     

Secondary design considerations for minimum bias estimation

Several authors have suggested the method of minimum bias estimation for estimating response surfaces. The minimum bias estimation procedure achieves minimum average squared bias of the fitted model without depending on the values of the unknown parameters of the true surface. The only requirement is that the design satisfies a simple estimability condition. Subject to providing minimum average squared bias, the minimum bias estimator also provides minimum average variance of ?(x) where ?(x) is the estimate of the response at the point x.

To support the estimation of the parameters in the fitted model, very little has been suggested in the way of experimental designs except to say that a full rank matrix X of independent variables should be used. This paper presents a closer look at the estimability conditions that are required for minimum bias estimation, and from the form of the matrix X, a formula is derived which measures the amount of design flexibility available. The design flexibility is termed “the degrees of freedom” of the X matrix and it is shown how the degrees of freedom can be used to decide if other design optimality criteria might be considered along with minimum bias estimation. Several examples are provided.     

Measuring the complexity of generalized linear hierarchical models

Measuring a statistical model's complexity is important for model criticism and comparison. However, it is unclear how to do this for hierarchical models due to uncertainty about how to count the random effects. The authors develop a complexity measure for generalized linear hierarchical models based on linear model theory. They demonstrate the new measure for binomial and Poisson observables modeled using various hierarchical structures, including a longitudinal model and an areal‐data model having both spatial clustering and pure heterogeneity random effects. They compare their new measure to a Bayesian index of model complexity, the effective number p_D of parameters (Spiegelhalter, Best, Carlin & van der Linde 2002); the comparisons are made in the binomial and Poisson cases via simulation and two real data examples. The two measures are usually close, but differ markedly in some instances where p_D is arguably inappropriate. Finally, the authors show how the new measure can be used to approach the difficult task of specifying prior distributions for variance components, and in the process cast further doubt on the commonly‐used vague inverse gamma prior.     

Sample size calculations for clinical studies allowing for uncertainty about the variance

One of the most important steps in the design of a pharmaceutical clinical trial is the estimation of the sample size. For a superiority trial the sample size formula (to achieve a stated power) would be based on a given clinically meaningful difference and a value for the population variance. The formula is typically used as though this population variance is known whereas in reality it is unknown and is replaced by an estimate with its associated uncertainty. The variance estimate would be derived from an earlier similarly designed study (or an overall estimate from several previous studies) and its precision would depend on its degrees of freedom. This paper provides a solution for the calculation of sample sizes that allows for the imprecision in the estimate of the sample variance and shows how traditional formulae give sample sizes that are too small since they do not allow for this uncertainty with the deficiency being more acute with fewer degrees of freedom. It is recommended that the methodology described in this paper should be used when the sample variance has less than 200 degrees of freedom.     

A method for choosing the smoothing parameter in a semi-parametric model for detecting change-points in blood flow

In a smoothing spline model with unknown change-points, the choice of the smoothing parameter strongly influences the estimation of the change-point locations and the function at the change-points. In a tumor biology example, where change-points in blood flow in response to treatment were of interest, choosing the smoothing parameter based on minimizing generalized cross-validation (GCV) gave unsatisfactory estimates of the change-points. We propose a new method, aGCV, that re-weights the residual sum of squares and generalized degrees of freedom terms from GCV. The weight is chosen to maximize the decrease in the generalized degrees of freedom as a function of the weight value, while simultaneously minimizing aGCV as a function of the smoothing parameter and the change-points. Compared with GCV, simulation studies suggest that the aGCV method yields improved estimates of the change-point and the value of the function at the change-point.     

Geometry and Degrees of Freedom of Linearly Constrained Generalized Lasso

The least squares fit in a linear regression is always unique even when the design matrix has rank deficiency. In this paper, we extend this classic result to linearly constrained generalized lasso. It is shown that under a mild condition, the fit can be represented as a projection onto a polytope and, hence, is unique no matter whether design matrix X has full column rank or not. Furthermore, a formula for the degrees of freedom is derived to characterize the effective number of parameters. It directly yields an unbiased estimate of degrees of freedom, which can be incorporated in an information criterion for model selection.     

Adaptive order selection for autoregressive models

Autoregressive model is a popular method for analysing the time dependent data, where selection of order parameter is imperative. Two commonly used selection criteria are the Akaike information criterion (AIC) and the Bayesian information criterion (BIC), which are known to suffer the potential problems regarding overfit and underfit, respectively. To our knowledge, there does not exist a criterion in the literature that can satisfactorily perform under various situations. Therefore, in this paper, we focus on forecasting the future values of an observed time series and propose an adaptive idea to combine the advantages of AIC and BIC but to mitigate their weaknesses based on the concept of generalized degrees of freedom. Instead of applying a fixed criterion to select the order parameter, we propose an approximately unbiased estimator of mean squared prediction errors based on a data perturbation technique for fairly comparing between AIC and BIC. Then use the selected criterion to determine the final order parameter. Some numerical experiments are performed to show the superiority of the proposed method and a real data set of the retail price index of China from 1952 to 2008 is also applied for illustration.     

Model selection for two-sample problems with right-censored data: An application of Cox model

For investigating differences between two treatment groups in medical science, selecting a suitable model to capture the underlying survival function for each group with some covariates is an important issue. Many methods, such as stratified Cox model and unstratified Cox model, have been proposed for investigating the problem. However, different models generally perform differently under different circumstances and none dominates the others. In this article, we focus on two sample problems with right-censored data and propose a model selection criterion based on an approximately unbiased estimator of Kullback-Leibler loss, which accounts for estimation uncertainty in estimated survival functions obtained by various candidate models. The effectiveness of the proposed method is justified by some simulation studies and it also applied to an HIV+ data set for illustration.     

A recommended analysis for 2 × 2 crossover trials with baseline measurements

In many two‐period, two‐treatment (2 × 2) crossover trials, for each subject, a continuous response of interest is measured before and after administration of the assigned treatment within each period. The resulting data are typically used to test a null hypothesis involving the true difference in treatment response means. We show that the power achieved by different statistical approaches is greatly influenced by (i) the ‘structure’ of the variance–covariance matrix of the vector of within‐subject responses and (ii) how the baseline (i.e., pre‐treatment) responses are accounted for in the analysis. For (ii), we compare different approaches including ignoring one or both period baselines, using a common change from baseline analysis (which we advise against), using functions of one or both baselines as period‐specific or period‐invariant covariates, and doing joint modeling of the post‐baseline and baseline responses with corresponding mean constraints for the latter. Based on theoretical arguments and simulation‐based type I error rate and power properties, we recommend an analysis of covariance approach that uses the within‐subject difference in treatment responses as the dependent variable and the corresponding difference in baseline responses as a covariate. Data from three clinical trials are used to illustrate the main points. Copyright © 2014 John Wiley & Sons, Ltd.     

Economical process adjustment with measurement error

Srivastava and Wu (1997) considered a random walk model with sampling interval and measurement error which was assumed to be white noise. In this paper, we consider the situation in which the measurement error is also a random walk. It is assumed that there is a sampling cost and an adjustment cost. The cost of deviating from the target value is assumed to be proportional to the square of the deviations. The long-run average cost rate is evaluated exactly in terms of the first four moments of a randomly stopped random walk. Using approximations of those moments, optimum, values of the control parameters are given.     

On an adjustment of degrees of freedom in the minimim mean squared error ertimator

In this paper, we consider an adjustment of degrees of freedom in the minimum mean squared error (MMSE) estimator, We derive the exact MSE of the adjusted MMSE (AMMSE) estimator, and compare the MSE of the AMMSE estimator with those of the Stein-(SR), positive-part Stein-rule (PSR) and MMSE estimators by numerical evaluations. It is shown that the adjustment of degrees of freedom is effective when the noncentrality parameter is close to zero, and the MSE performance of the MMSE estimator can be improved in the wide region of the noncentrality parameter by the adjustment, ft is also shown that the AMMSE estimator can have the smaller MSE than the PSR estimator in the wide region of the noncentrality parameter     

Estimation of a rare sensitive attribute in probability proportional to size measures using Poisson distribution

We propose new variants of Land et al.’s [Estimation of a rare sensitive attribute using Poisson distribution. Statistics. 2011. DOI: 10.1080/02331888.2010.524300] randomized response model when a population consists of some clusters and the population is stratified with some clusters in each stratum. The estimator for the mean number of persons who possess a rare sensitive attribute, its variance, and the variance estimator are suggested when the parameter of a rare unrelated attribute is assumed to be known and unknown. The clusters are selected with and without replacement. When they are selected with replacement, the selecting probabilities for each cluster are defined depending on the cluster sizes and with equal probability. In addition, the variance comparison between a probability proportional to size (PPS) and PPS for stratification are performed. When the parameters vary in clusters, the stratified PPS has better efficiency than the PPS.     

An improved Cp criterion for spline smoothing

Spline smoothing is a popular technique for curve fitting, in which selection of the smoothing parameter is crucial. Many methods such as Mallows’ C_p, generalized maximum likelihood (GML), and the extended exponential (EE) criterion have been proposed to select this parameter. Although C_p is shown to be asymptotically optimal, it is usually outperformed by other selection criteria for small to moderate sample sizes due to its high variability. On the other hand, GML and EE are more stable than C_p, but they do not possess the same asymptotic optimality as C_p. Instead of selecting this smoothing parameter directly using C_p, we propose to select among a small class of selection criteria based on Stein's unbiased risk estimate (SURE). Due to the selection effect, the spline estimate obtained from a criterion in this class is nonlinear. Thus, the effective degrees of freedom in SURE contains an adjustment term in addition to the trace of the smoothing matrix, which cannot be ignored in small to moderate sample sizes. The resulting criterion, which we call adaptive C_p, is shown to have an analytic expression, and hence can be efficiently computed. Moreover, adaptive C_p is not only demonstrated to be superior and more stable than commonly used selection criteria in a simulation study, but also shown to possess the same asymptotic optimality as C_p.     

A four-moment approach and other practical solutions to the behrens-f1sher problem

Fixed sample size approximately similar tests for the Behrens-Fisher problem are studied and compared with various other tests suggested in current sttistical methodelogy texts. Several fourmoment approxiamtely similar tests are developed and offered as alternatives. These tests are shown to be good practical solutions which are easily implemented in practice.