Journal Of Statistics Research Articles

Marginal regression analysis of recurrent events with coarsened censoring times.

Motivated by an ongoing pediatric mental health care (PMHC) study, this article presents weakly structured methods for analyzing doubly censored recurrent event data where only coarsened information on censoring is available. The study extracted administrative records of emergency department visits from provincial health administrative databases. The available information of each individual subject is limited to a subject-specific time window determined up to concealed data. To evaluate time-dependent effect of exposures, we adapt the local linear estimation with right censored survival times under the Cox regression model with time-varying coefficients (cf. Cai and Sun, Scandinavian Journal of Statistics 2003, 30, 93-111). We establish the pointwise consistency and asymptotic normality of the regression parameter estimator, and examine its performance by simulation. The PMHC study illustrates the proposed approach throughout the article.

Reviewer Acknowledgements for International Journal of Statistics and Probability, Vol. 5, No. 2

International Journal of Statistics and Probability wishes to acknowledge the following individuals for their assistance with peer review of manuscripts for this issue. Their help and contributions in maintaining the quality of the journal is greatly appreciated.Many authors, regardless of whether International Journal of Statistics and Probability publishes their work, appreciate the helpful feedback provided by the reviewers.Reviewers for Volume 5, Number 2Encarnación Alvarez-VerdejoGabriel A. OkyereGane Samb LoMarcelo BourguignonNicolas MARIEPhilip WestgateSajid AliSamah M. Abo-El-HadidSamir SafiShahedul KhanSohair F. HigaziWojciech GamrotZaixing Li Wendy SmithOn behalf of,The Editorial Board of International Journal of Statistics and ProbabilityCanadian Center of Science and Education

A sequential scaled pairwise selection approach to edge detection in nonparanormal graphical models

AbstractWe deal with the problem of edge detection in high‐dimensional nonparanormal graphical models. A nonparanormal graphical model is first transformed into a Gaussian graphical model. Then a sequential scaled pairwise selection (SSPS) method which we propose is applied to the transformed model. The SSPS method is a neighbourhood detection approach which makes use of conditional regression models. First the response vector in each individual conditional regression model is scaled, then the response vectors are pooled together to form a single model. The features in this single model are selected pairwise by a sequential procedure, which reflects the intrinsic symmetry of the edges. At each step of the procedure, the current residual vector is projected into the space spanned by each pair of columns of the design matrix which correspond to the symmetric edges, and the selected set of edges is then augmented by the pair with the largest projection norm. The extended BIC (EBIC) is used as the stopping rule for the sequential procedure. The selection consistency of the SSPS method is established. Simulation studies and the analysis of a real data set are carried out to compare the SSPS method with other existing methods. The simulation studies demonstrate that the SSPS method outperforms the other methods. In addition, the SSPS method is computationally more appealing. The Canadian Journal of Statistics 44: 25–43; 2016 © 2016 Statistical Society of Canada

A simplified version of the maximum information per time unit method in computerized adaptive testing.

In this article, we propose a simplified version of the maximum information per time unit method (MIT; Fan, Wang, Chang, & Douglas, Journal of Educational and Behavioral Statistics 37: 655-670, 2012), or MIT-S, for computerized adaptive testing. Unlike the original MIT method, the proposed MIT-S method does not require fitting a response time model to the individual-level response time data. It is also computationally efficient. The performance of the MIT-S method was compared against that of the maximum information (MI) method in terms of measurement precision, testing time saving, and item pool usage under various item response theory (IRT) models. The results indicated that when the underlying IRT model is the two- or three-parameter logistic model, the MIT-S method maintains measurement precision and saves testing time. It performs similarly to the MI method in exposure control; both result in highly skewed item exposure distributions, due to heavy reliance on the highly discriminating items. If the underlying model is the one-parameter logistic (1PL) model, the MIT-S method maintains the measurement precision and saves a considerable amount of testing time. However, its heavy reliance on time-saving items leads to a highly skewed item exposure distribution. This weakness can be ameliorated by using randomesque exposure control, which successfully balances the item pool usage. Overall, the MIT-S method with randomesque exposure control is recommended for achieving better testing efficiency while maintaining measurement precision and balanced item pool usage when the underlying IRT model is 1PL.

Combination of Forecasts across Estimation Windows: An Application to Air Travel Demand

This paper applies combining forecasts of air travel demand generated from the same model but over different estimation windows. The combination approach used resorts to Pesaran and Pick (Journal of Business Economics and Statistics 2011; 29: 307–318), but the empirical application is extended in several ways. The forecasts are based on a seasonal Box–Jenkins model (SARIMA), which is adequate to forecast monthly air travel demand with distinct seasonal patterns at the largest German airport: Frankfurt am Main. Furthermore, forecasts with forecast horizons from 1 to 12 months ahead, which are based on different average estimation windows, expanding windows and single rolling windows, are compared with baseline forecasts based on an expanding window of the observations after a structural break. The forecast exercise shows that the average window forecasts mostly outperform the alternative single window forecasts. Copyright © 2016 John Wiley & Sons, Ltd.

Back Cover

Journal Article Back Cover Get access Journal of Survey Statistics and Methodology, Volume 4, Issue 1, March 2016, Page NP, https://doi.org/10.1093/jssam/smw005 Published: 01 March 2016

Cover Page

Journal Article Cover Page Get access Journal of Survey Statistics and Methodology, Volume 4, Issue 1, March 2016, Page NP, https://doi.org/10.1093/jssam/smw006 Published: 01 March 2016

Contents Page

Journal Article Contents Page Get access Journal of Survey Statistics and Methodology, Volume 4, Issue 1, March 2016, Page NP, https://doi.org/10.1093/jssam/smw009 Published: 01 March 2016

Jackknife empirical likelihood for comparing two Gini indices

AbstractThe focus of this paper is to derive the jackknife empirical likelihood for the difference of two Gini indices. For independent data we propose a novel U‐statistic, which allows direct utilization of the jackknife empirical likelihood without involving a nuisance parameter. For paired data we established Wilks’ theorem for the profile likelihood after maximization over the nuisance parameter. Simulation studies show that our method is comparable to existing empirical likelihood methods in terms of coverage accuracy, but obtains much shorter intervals. The proposed methods are illustrated via analyzing a real data set.The Canadian Journal of Statistics44: 102–119; 2016 © 2016 Statistical Society of Canada

Sample‐size calculation for tests of homogeneity

AbstractMixture models are widely used to explain excessive variation in observations that is not captured by standard parametric models, and they lead to suggestive latent structures. The hypothetical latent structure often needs critical examination based on experimental data. It is therefore important to know the sample size needed to ensure a reasonable chance of success. We investigate this issue for the EM‐test and the test. They are shown to be asymptotically equivalent and have simple limiting distributions under two sets of local alternatives for commonly used mixture models. We obtain a simple sample‐size formula and an associated simulation‐based calibration procedure, and we demonstrate via data examples and simulation studies that they provide useful guidance for several common mixture models. The Canadian Journal of Statistics 44: 82–101; 2016 © 2016 Statistical Society of Canada

Honoring the Lion: A Festschrift for Jan de Leeuw

This special volume celebrates the 20th anniversary of the Journal of Statistical Software (JSS) and is a Festschrift for its founding editor Jan de Leeuw. Jan recently retired from his long-held position as founding chair of the Department of Statistics at the University of California, Los Angeles. The contributions to this special volume look back at some of his research interests and accomplishments during the half-century that he has been active in psychometrics and statistics. In this introduction, the guest editors also reminisce on their own first encounters with Jan, ten years ago. Since that time JSS has solidified its place as a leading journal of computational statistics, a fact that has a lot to do with Jan's stewardship. We include a brief history of JSS.

Extended Intervened Geometric Distribution

Here we develop an extended version of the modified intervened geometric distribution of Kumar and Sreeja (<i>The Aligarh Journal of Statistics, 2014</i>) and investigate some of its important statistical properties. Parameters of the distribution are estimated by various methods of estimation such as the method of factorial moments, the method of mixed moments and the method of maximum likelihood. The distribution has been fitted to a real life data set for illustrating its practical relevance.

Asymptotic normality of locally modelled regression estimator for functional data

We focus on the nonparametric regression of a scalar response on a functional explanatory variable. As an alternative to the well-known Nadaraya-Watson estimator for regression function in this framework, the locally modelled regression estimator performs very well [cf. [Barrientos-Marin, J., Ferraty, F., and Vieu, P. (2010), ‘Locally Modelled Regression and Functional Data’, Journal of Nonparametric Statistics, 22, 617–632]. In this paper, the asymptotic properties of locally modelled regression estimator for functional data are considered. The mean-squared convergence as well as asymptotic normality for the estimator are established. We also adapt the empirical likelihood method to construct the point-wise confidence intervals for the regression function and derive the Wilk's phenomenon for the empirical likelihood inference. Furthermore, a simulation study is presented to illustrate our theoretical results.

A component lasso

AbstractWe propose a new sparse regression method called the component lasso, based on a simple idea. The method uses the connected‐components structure of the sample covariance matrix to split the problem into smaller ones. It then applies the lasso to each subproblem separately, obtaining a coefficient vector for each one. Finally, it uses non‐negative least squares to recombine the different vectors into a single solution. This step is useful in selecting and reweighting components that are correlated with the response. We prove that the component lasso is strongly sign consistent in a block‐diagonal setting. Simulated and real data examples show that the component lasso can outperform standard regression methods such as the lasso and elastic net, achieving a lower mean squared error as well as better support recovery. The modular structure of the algorithm also lends itself naturally to parallel computation. The Canadian Journal of Statistics 43: 624–646; 2015 © 2015 Statistical Society of Canada

Edgeworth expansions for two‐stage sampling with applications to stratified and cluster sampling

AbstractA two‐term Edgeworth expansion for the standardized version of the sample total in a two‐stage sampling design is derived. In particular, for the commonly used stratified and cluster sampling schemes, formal two‐term asymptotic expansions are obtained for the Studentized versions of the sample total. These results are applied in conjunction with the bootstrap to construct more accurate confidence intervals for the unknown population total in such sampling schemes. The Canadian Journal of Statistics 43: 578–599; 2015 © 2015 The Authors. The Canadian Journal of Statistics published by Wiley Periodicals, Inc. on behalf of Statistical Society of Canada

Subscription Page

Journal Article Subscription Page Get access Journal of Survey Statistics and Methodology, Volume 3, Issue 4, December 2015, Page NP, https://doi.org/10.1093/jssam/smv045 Published: 01 December 2015

Back Cover

Journal Article Back Cover Get access Journal of Survey Statistics and Methodology, Volume 3, Issue 4, December 2015, Page NP, https://doi.org/10.1093/jssam/smv042 Published: 01 December 2015

Acknowledgement of the Reviewers

Journal Article Acknowledgement of the Reviewers Get access Journal of Survey Statistics and Methodology, Volume 3, Issue 4, December 2015, Pages 566–568, https://doi.org/10.1093/jssam/smv041 Published: 24 November 2015

Semiparametric maximum likelihood estimation for a two‐sample density ratio model with right‐censored data

AbstractIn this paper we investigate a broader semiparametric two‐sample density ratio model based on two groups of right‐censored data. A semiparametric maximum likelihood estimator for the unknown finite and infinite dimensional parameters of the model is proposed and obtained by an EM algorithm. By using empirical process theory, we establish the uniform consistency and asymptotic normality of the proposed estimator. We moreover employ a Kolmogorov–Smirnov type test statistic to evaluate the model validity and a likelihood ratio test statistic to examine the treatment effects between the two groups. Simulation studies are conducted to assess the finite sample performance of the proposed estimator and to compare it with its alternatives. Finally a real data example is analyzed to illustrate its application. The Canadian Journal of Statistics 44: 58–81; 2016 © 2015 Statistical Society of Canada

Replicating the Results in ‘A New Model of Trend Inflation’ Using Particle Markov Chain Monte Carlo

SummaryAn article by Chan et al. () published in the Journal of Business and Economic Statistics introduces a new model for trend inflation. They allow the trend inflation to evolve according to a bounded random walk. In order to draw the latent states from their respective conditional posteriors, they use accept–reject Metropolis–Hastings procedures. We reproduce their results using particle Markov chain Monte Carlo (PMCMC), which approaches drawing the latent states from a different technical point of view by relying on combining Markov chain Monte Carlo and sequential Monte Carlo methods. To conclude: we are able to reproduce the results of Chan et al. (). Copyright © 2015 John Wiley &amp; Sons, Ltd.