Computational Statistics

Biogeography of fruit bats in Southeast Asia. Les S. Hall, Gordon G. Grigg, Craig Moritz, Besar Ketol, Isa Sait, Wahab Marni and M.T. Abdullah.

by M.T. Abdullah, PhD

Should read Abdullah (2003).

We studied on the biogeography and diversity of fruit bats in Southeast Asia, from Borneo to Asian Mainland.The... more We studied on the biogeography and diversity of fruit bats in Southeast Asia, from Borneo to Asian Mainland.The patterns of distribution of distribution, diversity and abundance are related to ecological and biogeographical factors and possible past Pleistocene events.

Les S. Hall, Gordon G. Grigg, Craig Moritz, Besar Ketol, Isa Sait, Wahab Marni and M.T. Abdullah. 2004. Sarawak Museum Journal 81: 191-284.

Morphometrical Variations of Malaysian Hipposideros Species. 2012

by M.T. Abdullah, PhD

Read Vijaya et al (2012)

A study on the morphometrical variations among four Malaysian Hipposideros species was conducted using voucher... more A study on the morphometrical variations among four Malaysian Hipposideros species was conducted using voucher specimens deposited in Universiti Malaysia Sarawak (UNIMAS) Zoological Museum and the Department of Widlife and National Park (DWNP) Kuala Lumpur. Twenty two individuals from four species of Hipposideros ater, H. bicolor, H. cineraceus and H. dyacorum were morphologically measured, in which a total of 27 linear parameters of body, skull and dentals of each were appropriately recorded. The statistical data were later subjected to discriminant function analysis (DFA) and canonical variate analysis (CVA) using SPSS version 15.0 and unweighted pair-group method average (UPGMA) cluster analysis using Minitab version 14.4. The highest character loadings observed in Function l, Function 2 and Function 3 were the forearm length (FA), the third digit second phalanx length (D3P2L) and the palatal length (PL) with standardised canonical discriminant function coefficient values of 21.910, 5.770 and 5.095, respectively. These three characters were identified as the best diagnostic features for discriminating these closely related species of Hipposideros. Hence, this morphometric approach could be a promising tool as an alternative to the molecular
DNA analysis for identification of Chiroptera species.

A Predictive Model to Differentiate the Fruit Bats Cynopterus brachyotis and C. cf. brachyotis Forest (Chiroptera: Pteropodidae) from Malaysia Using Multivariate Analysis. 2012

by M.T. Abdullah, PhD

Read also Abdullah (2003)

Field discrimination of Cynopterus brachyotis and C. cf. brachyotis Forest (as designated by Francis 2008) in southern... more Field discrimination of Cynopterus brachyotis and C. cf. brachyotis Forest (as designated by Francis 2008) in southern Thailand, Peninsular Malaysia, and Borneo is problematic. These 2 forms are sympatric in this region but are confined to different habitat types: C. brachyotis inhabits open habitats, orchards, and agricultural areas, while C. cf. brachyotis Forest is confined to primary and old secondary forests. In this study, we attempted to develop prediction models to identify both C. brachyotis and C. cf. brachyotis Forest in this region based on multivariate statistics. Two predictive models were generated using a canonical discriminant function, and it was found that 5 characters can be used to accurately identify museum vouchers of C. brachyotis and C. cf. brachyotis Forest. Four characters are needed for field identification of these 2 forms of Cynopterus in southern Thailand, Peninsular Malaysia, and Borneo. A review of the current taxonomy and classification indicated that there is a need to describe the 6 existing forms of the C. brachyotis complex in the Indo-Malayan region. This will aid conservationists, field ecologists, and taxonomists in taxonomic- and conservation-related decisions about this species complex.

EstCRM: An R package for Samejima's Continuous IRT Model.

by Cengiz Zopluoglu

http://apm.sagepub.com/content/current

EstCRM: An R Package for Samejima’s Continuous IRT Model, Applied Psychological Measurement, March 2012, 36, 149-150

Continuous Response Model (CRM) is an IRT model developed for continuous outcomes. CRM is not commonly used in... more Continuous Response Model (CRM) is an IRT model developed for continuous outcomes. CRM is not commonly used in practice although it is as old as the other well known and popular binary and polytomous IRT models. This may be due to the lack of an accessible software to estimate the model parameters. The R package, EstCRM, was developed to estimate the model parameters for the CRM.

Automatic repeated-loess decomposition of data consisting of sums of oscillatory curves

by David Foster

Repeated loess is a nonparametric procedure that uses progressive smoothing and differencing to decompose data... more Repeated loess is a nonparametric procedure that uses progressive smoothing and differencing to decompose data consisting of sums of curves. Smoothing is by locally weighted polynomial regression. Here the procedure was developed so that the decomposition into components was controlled automatically by the number of maxima in each component. The level of smoothing of each component was chosen to maximize the estimated probability of the observed number of maxima. No assumptions were made about the periodicity of components and only very weak assumptions about their shapes. The automatic procedure was applied to simulated data and to experimental data on human visual sensitivity to line orientation.

Random imperfection fields to model the size effect in laboratory wood specimens

by Sara Casciati

Casciati S. and Domaneschi M. (2007). “Random imperfection fields to model the size effect in laboratory wood specimens”. Structural Safety, 29(4), 308-321. ISSN: 0167-4730.

DATA E LUOGO DI PUBBLICAZIONE: October 2007; Elsevier Science Bv, 1000 AE Amsterdam, Netherlands.

ABSTRACT. The composite nature of a wood continuum prevents one from extrapolating the results of laboratory tests on... more ABSTRACT. The composite nature of a wood continuum prevents one from extrapolating the results of laboratory tests on standard wood specimens to structural elements of significant size. Therefore, these elements are usually tested under standardized loading conditions in order to detect a sort of average material behaviour.
In this paper, the initial step consists, instead, of testing the material specimens. The extension of the results to structural elements is then pursued by introducing a random field, or, in a discretized model, a random array of imperfections.
The calibration of the suitable spatial distribution of the imperfections is then investigated by a mixed experimental–numerical approach, for a reference beam. The analyses on the relative finite elements model are iterated to match the response of the full scale laboratory tests.

KEYWORDS: Biaxial tests; Finite element model; Imperfections; Laboratory tests; Random field; Wood specimens

Cohesive Crack Propagation in a Random Elastic Medium

by Sara Casciati

Bruggi M., Casciati S., and Faravelli L. (2008). “Cohesive crack propagation in a random elastic medium”. Probabilistic Engineering Mechanics, 23(1), 23-35. ISSN: 0266-8920.

DATA E LUOGO DI PUBBLICAZIONE: January 2008; Elsevier Sci Ltd, Kidlington, Oxford OX5 1GB, Oxon, England.

ABSTRACT. The issue of generating non-Gaussian, multivariate and correlated random fields, while preserving the... more ABSTRACT. The issue of generating non-Gaussian, multivariate and correlated random fields, while preserving the internal auto-correlation structure of each single-parameter field, is discussed with reference to the problem of cohesive crack propagation. Three different fields are introduced to model the spatial variability of the Young modulus, the tensile strength of the material, and the fracture energy, respectively. Within a finite-element context, the crack-propagation phenomenon is analyzed by coupling a Monte Carlo simulation scheme with an iterative solution algorithm based on a truly-mixed variational formulation which is derived from the Hellinger–Reissner principle. The selected approach presents the advantage of exploiting the finite-element technology without the need to introduce additional modes to model the displacement discontinuity along the crack boundaries. Furthermore, the accuracy of the stress estimate pursued by the truly-mixed approach is highly desirable, the direction of crack propagation being determined on the basis of the principal stress criterion. The numerical example of a plain concrete beam with initial crack under a three-point bending test is considered. The statistics of the response is analyzed in terms of peak load and load–mid deflection curves, in order to investigate the effects of the uncertainties on both the carrying capacity and the post-peak behaviour. A sensitivity analysis is preliminarily performed and its results emphasize the negative effects of not accounting for the auto-correlation structure of each random field. A probabilistic method is then applied to enforce the auto-correlation without significantly altering the target marginal distributions. The novelty of the proposed approach with respect to other methods found in the literature consists of not requiring the a priori knowledge of the global correlation structure of the multivariate random field.

KEYWORDS: Multivariate non-Gaussian random fields; Auto-correlation; Cohesive crack propagation; Truly-mixed finite element method; Monte Carlo simulations

Multi-level Bootstrapping the SAS® Way

by John Eustaquio

Draft Only

Bootstrapping is a non-parametric method since specific assumptions are not done about the distribution from which the... more Bootstrapping is a non-parametric method since specific assumptions are not done about the distribution from which the data arise. Also, it is widely used for it can approximate the entire sampling distribution of some estimator; can estimate the bias, or standard error, or a confidence interval of the parameter. The problem of the classical bootstrapping procedure is that it always gives the value of a statistics with high standard errors. Multi-level bootstrapping is proposed as the extended version of bootstrapping. It estimates the parameter of interest from bootstrap resamples giving you a smaller variability or higher reliability as compared to the original bootstrap. To evaluate the multi-level bootstrapping, three samples were generated from a Gaussian distribution; one is for the small sample case, the second for a large sample case and the other is for the unrepresentative sample case. The multi-level bootstrapping is suitable in determining the distribution of a sample when the goal is to achieve low standard errors. For both small and large samples, the multi-level bootstrapping procedure yielded lower standard errors than the classical bootstrapping procedure. In cases where there is an unrepresentative sample, the multi-level bootstrapping procedure produced low standard errors although the estimates are highly biased.

Multi-level Bootstrapping the SAS® Way

by John Eustaquio

Draft Only

Bootstrapping is a non-parametric method since specific assumptions are not done about the distribution from which the... more Bootstrapping is a non-parametric method since specific assumptions are not done about the distribution from which the data arise. Also, it is widely used for it can approximate the entire sampling distribution of some estimator; can estimate the bias, or standard error, or a confidence interval of the parameter. The problem of the classical bootstrapping procedure is that it always gives the value of a statistics with high standard errors. Multi-level bootstrapping is proposed as the extended version of bootstrapping. It estimates the parameter of interest from bootstrap resamples giving you a smaller variability or higher reliability as compared to the original bootstrap. To evaluate the multi-level bootstrapping, three samples were generated from a Gaussian distribution; one is for the small sample case, the second for a large sample case and the other is for the unrepresentative sample case. The multi-level bootstrapping is suitable in determining the distribution of a sample when the goal is to achieve low standard errors. For both small and large samples, the multi-level bootstrapping procedure yielded lower standard errors than the classical bootstrapping procedure. In cases where there is an unrepresentative sample, the multi-level bootstrapping procedure produced low standard errors although the estimates are highly biased.

Unimodal regression via prefix isotonic regression

by Quentin F. Stout

appears in Computational Statistics and Data Analysis 53 (2008), pp. 289–297

Optimal few-stage designs for clinical trials

by Quentin F. Stout

Optimal screening designs with flexible cost structures

by Quentin F. Stout

Adaptive Designs for Balancing Complex Objectives

by Quentin F. Stout

Bandit strategies for ethical sequential allocation

by Quentin F. Stout

Optimal adaptive equal allocation rules

by Quentin F. Stout

Optimal allocation for estimating the mean of a bivariate polynomial

by Quentin F. Stout

Adaptive Allocation in the Presence of Missing Observations

by Quentin F. Stout

Sequentially deciding between two experiments for estimating a common success probability

by Quentin F. Stout

Using path induction to evaluate sequential allocation procedures

by Quentin F. Stout

Optimal few-stage designs

by Quentin F. Stout