Multiple approaches have been developed for improving predictive performance of a system by creating and combining various learned models. There are two main approaches to creating model ensembles. The first is to create a set of learned models by applying an algorithm repeatedly to different training sample data, the second applies various learning algorithms to the same sample data. The predi...

Extensive empirical experience suggests that traditional forecasting approaches are subject to more or less severe model misspecifications which affect true (out-of-sample) oneas well as multi-step ahead forecasting performances. The main problems are due to non-stationarity and non-Gaussianity. In order to overcome these difficulties, we propose a prototypical design derived from a traditional...

Background Cluster randomized trials have been utilized to evaluate the effectiveness of HIV prevention strategies on reducing incidence. Design of such studies must take into account possible correlation of outcomes within randomized units. Purpose To discuss power and sample size considerations for cluster randomized trials of combination HIV prevention, using an HIV prevention study in Botsw...

In this paper, we propose a new algorithm for automated image registration or precise correction of satellite images. We assume that ground control points used previously are stored within the system. The algorithm first applies matching between the GCP chips stored and a new image to be registered and creates new control points. An automated stereo matching based on normalized cross correlatio...

The complexity of tissue and cell proteomes and the vast dynamic range of protein abundance present a formidable challenge for analysis that no one analytical technique can overcome. As a result, there is a need to integrate technologies to achieve the high-resolution and high-sensitivity analysis of complex biological samples. The combined technologies of separation science and biological mass...

In clinical research, parameters required for sample size calculation are usually unknown. A typical approach is to use estimates from some pilot studies as the true parameters in the calculation. This approach, however, does not take into consideration sampling error. Thus, the resulting sample size could be misleading if the sampling error is substantial. As an alternative, we suggest a Bayes...

We consider the problem of learning an unknown (overcomplete) basis from an unknown sparse linear combination. Introducing the “sparse coding neural gas” algorithm, we show how to employ a combination of the original neural gas algorithm and Oja’s rule in order to learn a simple sparse code that represents each training sample by a multiple of one basis vector. We generalise this algorithm usin...

BACKGROUND In certain diseases clinical experts may judge that the intervention with the best prospects is the addition of two treatments to the standard of care. This can either be tested with a simple randomized trial of combination versus standard treatment or with a 2 x 2 factorial design. METHODS We compared the two approaches using the design of a new trial in tuberculous meningitis as ...

We explore the extension of James-Stein type estimators in a direction that enables them to preserve their superiority when the sample size goes to infinity. Instead of shrinking a base estimator towards a fixed point, we shrink it towards a data-dependent point. We provide an analytic expression for the asymptotic risk and bias of James-Stein type estimators shrunk towards a data-dependent poi...