On almost sure convergence for sums of stochastic sequence

In this work we discuss almost sure convergence for sums of arbitrarily dependent stochastic sequence under different conditions of Chung’s type. Our approach is based on the stopping time technique and the theorem of convergence for martingale difference sequence. Meanwhile, the results here include some relevant classical conclusions.     

Unbalanced cluster sizes and rates of convergence in mixed-effects models for clustered data

ABSTRACT

Convergence problems often arise when complex linear mixed-effects models are fitted. Previous simulation studies (see, e.g. [Buyse M, Molenberghs G, Burzykowski T, Renard D, Geys H. The validation of surrogate endpoints in meta-analyses of randomized experiments. Biostatistics. 2000;1:49–67, Renard D, Geys H, Molenberghs G, Burzykowski T, Buyse M. Validation of surrogate endpoints in multiple randomized clinical trials with discrete outcomes. Biom J. 2002;44:921–935]) have shown that model convergence rates were higher (i) when the number of available clusters in the data increased, and (ii) when the size of the between-cluster variability increased (relative to the size of the residual variability). The aim of the present simulation study is to further extend these findings by examining the effect of an additional factor that is hypothesized to affect model convergence, i.e. imbalance in cluster size. The results showed that divergence rates were substantially higher for data sets with unbalanced cluster sizes – in particular when the model at hand had a complex hierarchical structure. Furthermore, the use of multiple imputation to restore ‘balance’ in unbalanced data sets reduces model convergence problems.     

Bayesian alignment of proteins via Delaunay tetrahedralization

An active area of research in bioinformatics is finding structural similarity of proteins by alignment. Among many methods, the popular one is to find the similarity based on statistical features. This method involves gathering information from the complex biomolecule structure and obtaining the best alignment by maximizing the number of matched features. In this paper, after reviewing statistical models for matching the structural biomolecule, it is shown that local alignment based on the Delaunay tetrahedralization (DT) can be used for Bayesian alignment of proteins. In this method, we use DT to add a priori structural information of protein in the Bayesian methodology. We demonstrate that this method shows advantages over competing methods in achieving a global alignment of proteins, accelerating the convergence rate and improving the parameter estimates.     

Asymptotic convergence in distribution for spearman's rank correlation coefficient

The Edgeworth expansion for the distribution function of Spearman's rank correlation coefficient may be used to show that the rates of convergence for the normal and Pearson type II approximations are l/nand l/n² respectively. Using the Edgeworth expansion up to terms involving the sixth moment of the exact distribution allows an approximation with an error of order l/n³.     

Toeplitz lemma,complete convergence,and complete moment convergence

In this paper, we study the Toeplitz lemma, the Cesàro mean convergence theorem, and the Kronecker lemma. At first, we study “complete convergence” versions of the Toeplitz lemma, the Cesàro mean convergence theorem, and the Kronecker lemma. Two counterexamples show that they can fail in general and some sufficient conditions for “complete convergence” version of the Cesàro mean convergence theorem are given. Second, we introduce two classes of complete moment convergence, which are stronger versions of mean convergence and consider the Toeplitz lemma, the Cesàro mean convergence theorem, and the Kronecker lemma under these two classes of complete moment convergence.     

Diffusion approximations for insurance risk processes

We consider a risk-reserve process for an insurance company where premium income and the claim sum process are modeled as a renewal reward processes. Moreover, dividends are paid out according to a barrier rule. The aim of the article is to establish a diffusion approximation of this model and to compute ruin probabilities (in finite and in infinite time) and other relevant statistics approximately using the limiting diffusion process. We also demonstrate that, under special circumstances, there exists a stationary distribution for the limiting diffusion.     

Maximum-likelihood-schätzungen für parameter in homogenen markovschen ketten

A class of homogeneous MARKOV chains with arbitrary intial distributions is considered, which in general contains also the class treated by BILLINGSLEY [2]. The existence of strongly consistent and asymptotically normal maximum-likelihood estimates for unknown parameters is shown. Also the case of a random sample size is treated.     

Berry esséen type bounds for nonlinear models

Berry Esséen type bounds for the distribution of the nonlinear least-squares estimator as well as of a certain statistic useful for testing a hypothesis on a subset of the parameter vector are derived.     

Controlling Type-I Error Rate in Monitoring Structural Changes Using Partially Sequential Procedures

In the present article, we develop some asymptotically power on partially sequential nonparametric tests for monitoring structural changes. Our test procedures are based on Wilcoxon score. We use the idea of curved stopping boundaries. We derive some exact results and perform simulation studies to provide various properties of the tests. We see that one of the proposed procedures significantly controls the Type I error rate. This procedure may be very effective for fluctuation monitoring. We illustrate the procedures by using real life data from the stock market.     

Parameter change test for zero-inflated generalized Poisson autoregressive models

In this paper, we consider the problem of testing for parameter change in zero-inflated generalized Poisson (ZIGP) autoregressive models. We verify that the ZIGP process is stationary and ergodic and that the conditional maximum likelihood estimator (CMLE) is strongly consistent and asymptotically normal. Based on these results, we construct CMLE- and residual-based cumulative sum tests and show that their limiting null distributions are a function of independent Brownian bridges. The simulation results are provided for illustration. A real data analysis is performed on some crime data of Australia.