Electronic marking of statistics assessments for bioscience students

We describe the main features of a program written to perform electronic marking of quantitative or simple text questions. One of the main benefits is that it can check answers for being consistent with earlier errors, so can cope with a range of numerical questions. We summarise our experience of using it in a statistics course taught to 200 bioscience students.

A covariate adjustment for zero-truncated approaches to estimating the size of hidden and elusive populations

In this paper we consider the estimation of population size from onesource capture–recapture data, that is, a list in which individuals can potentially be found repeatedly and where the question is how many individuals are missed by the list. As a typical example, we provide data from a drug user study in Bangkok from 2001 where the list consists of drug users who repeatedly contact treatment institutions. Drug users with 1, 2, 3, . . . contacts occur, but drug users with zero contacts are not present, requiring the size of this group to be estimated. Statistically, these data can be considered as stemming from a zero-truncated count distribution.We revisit an estimator for the population size suggested by Zelterman that is known to be robust under potential unobserved heterogeneity. We demonstrate that the Zelterman estimator can be viewed as a maximum likelihood estimator for a locally truncated Poisson likelihood which is equivalent to a binomial likelihood. This result allows the extension of the Zelterman estimator by means of logistic regression to include observed heterogeneity in the form of covariates. We also review an estimator proposed by Chao and explain why we are not able to obtain similar results for this estimator. The Zelterman estimator is applied in two case studies, the first a drug user study from Bangkok, the second an illegal immigrant study in the Netherlands. Our results suggest the new estimator should be used, in particular, if substantial unobserved heterogeneity is present.

The binding site distance test score: a robust method for the assessment of predicted protein binding sites

We propose a novel method for scoring the accuracy of protein binding site predictions – the Binding-site Distance Test (BDT) score. Recently, the Matthews Correlation Coefficient (MCC) has been used to evaluate binding site predictions, both by developers of new methods and by the assessors for the community wide prediction experiment – CASP8. Whilst being a rigorous scoring method, the MCC does not take into account the actual 3D location of the predicted residues from the observed binding site. Thus, an incorrectly predicted site that is nevertheless close to the observed binding site will obtain an identical score to the same number of nonbinding residues predicted at random. The MCC is somewhat affected by the subjectivity of determining observed binding residues and the ambiguity of choosing distance cutoffs. By contrast the BDT method produces continuous scores ranging between 0 and 1, relating to the distance between the predicted and observed residues. Residues predicted close to the binding site will score higher than those more distant, providing a better reflection of the true accuracy of predictions. The CASP8 function predictions were evaluated using both the MCC and BDT methods and the scores were compared. The BDT was found to strongly correlate with the MCC scores whilst also being less susceptible to the subjectivity of defining binding residues. We therefore suggest that this new simple score is a potentially more robust method for future evaluations of protein-ligand binding site predictions.

The sequential analysis of repeated binary responses: a score test for the case of three time points

In this paper a robust method is developed for the analysis of data consisting of repeated binary observations taken at up to three fixed time points on each subject. The primary objective is to compare outcomes at the last time point, using earlier observations to predict this for subjects with incomplete records. A score test is derived. The method is developed for application to sequential clinical trials, as at interim analyses there will be many incomplete records occurring in non-informative patterns. Motivation for the methodology comes from experience with clinical trials in stroke and head injury, and data from one such trial is used to illustrate the approach. Extensions to more than three time points and to allow for stratification are discussed. Copyright © 2005 John Wiley & Sons, Ltd.

Extending Zelterman’s approach for robust estimation of population size to zero-truncated clustered data

Estimation of population size with missing zero-class is an important problem that is encountered in epidemiological assessment studies. Fitting a Poisson model to the observed data by the method of maximum likelihood and estimation of the population size based on this fit is an approach that has been widely used for this purpose. In practice, however, the Poisson assumption is seldom satisfied. Zelterman (1988) has proposed a robust estimator for unclustered data that works well in a wide class of distributions applicable for count data. In the work presented here, we extend this estimator to clustered data. The estimator requires fitting a zero-truncated homogeneous Poisson model by maximum likelihood and thereby using a Horvitz-Thompson estimator of population size. This was found to work well, when the data follow the hypothesized homogeneous Poisson model. However, when the true distribution deviates from the hypothesized model, the population size was found to be underestimated. In the search of a more robust estimator, we focused on three models that use all clusters with exactly one case, those clusters with exactly two cases and those with exactly three cases to estimate the probability of the zero-class and thereby use data collected on all the clusters in the Horvitz-Thompson estimator of population size. Loss in efficiency associated with gain in robustness was examined based on a simulation study. As a trade-off between gain in robustness and loss in efficiency, the model that uses data collected on clusters with at most three cases to estimate the probability of the zero-class was found to be preferred in general. In applications, we recommend obtaining estimates from all three models and making a choice considering the estimates from the three models, robustness and the loss in efficiency. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

The best of both worlds: Producing national statistics using participatory methods

This paper presents our experience with combining statistical principles and participatory methods to generate national statistics. The methodology was developed in Malawi during 1999–2002. We demonstrate that if PRA is combined with statistical principles (including probability-based sampling and standardization), it can produce total population statistics and estimates of the proportion of households with certain characteristics (e.g., poverty). It can also provide quantitative data on complex issues of national importance such as poverty targeting. This approach is distinct from previous PRA-based approaches, which generate numbers at community level but only provide qualitative information at national level.

Some new three-level orthogonal main effects plans robust to model uncertainty

In this paper, we list some new orthogonal main effects plans for three-level designs for 4, 5 and 6 factors in IS runs and compare them with designs obtained from the existing L-18 orthogonal array. We show that these new designs have better projection properties and can provide better parameter estimates for a range of possible models. Additionally, we study designs in other smaller run-sizes when there are insufficient resources to perform an 18-run experiment. Plans for three-level designs for 4, 5 and 6 factors in 13 to 17 runs axe given. We show that the best designs here are efficient and deserve strong consideration in many practical situations.

Robust identification for linear-in-the-parameters models

In this paper new robust nonlinear model construction algorithms for a large class of linear-in-the-parameters models are introduced to enhance model robustness, including three algorithms using combined A- or D-optimality or PRESS statistic (Predicted REsidual Sum of Squares) with regularised orthogonal least squares algorithm respectively. A common characteristic of these algorithms is that the inherent computation efficiency associated with the orthogonalisation scheme in orthogonal least squares or regularised orthogonal least squares has been extended such that the new algorithms are computationally efficient. A numerical example is included to demonstrate effectiveness of the algorithms. Copyright (C) 2003 IFAC.