Robust Mixture Modelling Using the t Distribution

Normal mixture models are being increasingly used to model the distributions of a wide variety of random phenomena and to cluster sets of continuous multivariate data. However, for a set of data containing a group or groups of observations with longer than normal tails or atypical observations, the use of normal components may unduly affect the fit of the mixture model. In this paper, we consider a more robust approach by modelling the data by a mixture of t distributions. The use of the ECM algorithm to fit this t mixture model is described and examples of its use are given in the context of clustering multivariate data in the presence of atypical observations in the form of background noise.

Mixture models for capture-recapture count data

The contribution investigates the problem of estimating the size of a population, also known as the missing cases problem. Suppose a registration system is targeting to identify all cases having a certain characteristic such as a specific disease (cancer, heart disease, ...), disease related condition (HIV, heroin use, ...) or a specific behavior (driving a car without license). Every case in such a registration system has a certain notification history in that it might have been identified several times (at least once) which can be understood as a particular capture-recapture situation. Typically, cases are left out which have never been listed at any occasion, and it is this frequency one wants to estimate. In this paper modelling is concentrating on the counting distribution, e.g. the distribution of the variable that counts how often a given case has been identified by the registration system. Besides very simple models like the binomial or Poisson distribution, finite (nonparametric) mixtures of these are considered providing rather flexible modelling tools. Estimation is done using maximum likelihood by means of the EM algorithm. A case study on heroin users in Bangkok in the year 2001 is completing the contribution.

Parametric and nonparametric statistical modelling of crop yield: implications for pricing crop insurance contracts

This article considers alternative methods to calculate the fair premium rate of crop insurance contracts based on county yields. The premium rate was calculated using parametric and nonparametric approaches to estimate the conditional agricultural yield density. These methods were applied to a data set of county yield provided by the Statistical and Geography Brazilian Institute (IBGE), for the period of 1990 through 2002, for soybean, corn and wheat, in the State of Paran. In this article, we propose methodological alternatives to pricing crop insurance contracts resulting in more accurate premium rates in a situation of limited data.

Finite mixture analysis of beauty-contest data using generalised beta distributions

This paper introduces a mixture model based on the beta distribution, without preestablishedmeans and variances, to analyze a large set of Beauty-Contest data obtainedfrom diverse groups of experiments (Bosch-Domenech et al. 2002). This model gives a bettert of the experimental data, and more precision to the hypothesis that a large proportionof individuals follow a common pattern of reasoning, described as iterated best reply (degenerate),than mixture models based on the normal distribution. The analysis shows thatthe means of the distributions across the groups of experiments are pretty stable, while theproportions of choices at dierent levels of reasoning vary across groups.

Statistical modelling of CSP solving algorithms performance

The goal of this work is to try to create a statistical model, based only on easily computable parameters from the CSP problem to predict runtime behaviour of the solving algorithms, and let us choose the best algorithm to solve the problem. Although it seems that the obvious choice should be MAC, experimental results obtained so far show, that with big numbers of variables, other algorithms perfom much better, specially for hard problems in the transition phase.

Technical Note: Dissolved organic matter fluorescence a finite mixture approach to deconvolve excitation-emission matrices

The analysis of the shape of excitation-emission matrices (EEMs) is a relevant tool for exploring the origin, transport and fate of dissolved organic matter (DOM) in aquatic ecosystems. Within this context, the decomposition of EEMs is acquiring a notable relevance. A simple mathematical algorithm that automatically deconvolves individual EEMs is described, creating new possibilities for the comparison of DOM fluorescence properties and EEMs that are very different from each other. A mixture model approach is adopted to decompose complex surfaces into sub-peaks. The laplacian operator and the Nelder-Mead optimisation algorithm are implemented to individuate and automatically locate potential peaks in the EEM landscape. The EEMs of a simple artificial mixture of fluorophores and DOM samples collected in a Mediterranean river are used to describe the model application and to illustrate a strategy that optimises the search for the optimal output.

On Finite Mixture of Pareto and Beta Distnbutions

Department of Statistics, Cochin University of Science and Technology

Intro to Statistical Modelling

Lecture notes for a first year statistical modelling course.

Statistical modelling of grapevine yield in the Port Wine region under present and future climate conditions

The impact of projected climate change on wine production was analysed for the Demarcated Region of Douro, Portugal. A statistical grapevine yield model (GYM) was developed using climate parameters as predictors. Statistically significant correlations were identified between annual yield and monthly mean temperatures and monthly precipitation totals during the growing cycle. These atmospheric factors control grapevine yield in the region, with the GYM explaining 50.4% of the total variance in the yield time series in recent decades. Anomalously high March rainfall (during budburst, shoot and inflorescence development) favours yield, as well as anomalously high temperatures and low precipitation amounts in May and June (May: flowering and June: berry development). The GYM was applied to a regional climate model output, which was shown to realistically reproduce the GYM predictors. Finally, using ensemble simulations under the A1B emission scenario, projections for GYM-derived yield in the Douro Region, and for the whole of the twenty-first century, were analysed. A slight upward trend in yield is projected to occur until about 2050, followed by a steep and continuous increase until the end of the twenty-first century, when yield is projected to be about 800 kg/ha above current values. While this estimate is based on meteorological parameters alone, changes due to elevated CO2 may further enhance this effect. In spite of the associated uncertainties, it can be stated that projected climate change may significantly benefit wine yield in the Douro Valley.

Statistical modelling of extreme rainfall in Madeira Island

Extreme rainfall events have triggered a significant number of flash floods in Madeira Island along its past and recent history. Madeira is a volcanic island where the spatial rainfall distribution is strongly affected by its rugged topography. In this thesis, annual maximum of daily rainfall data from 25 rain gauge stations located in Madeira Island were modelled by the generalised extreme value distribution. Also, the hypothesis of a Gumbel distribution was tested by two methods and the existence of a linear trend in both distributions parameters was analysed. Estimates for the 50– and 100–year return levels were also obtained. Still in an univariate context, the assumption that a distribution function belongs to the domain of attraction of an extreme value distribution for monthly maximum rainfall data was tested for the rainy season. The available data was then analysed in order to find the most suitable domain of attraction for the sampled distribution. In a different approach, a search for thresholds was also performed for daily rainfall values through a graphical analysis. In a multivariate context, a study was made on the dependence between extreme rainfall values from the considered stations based on Kendall’s τ measure. This study suggests the influence of factors such as altitude, slope orientation, distance between stations and their proximity of the sea on the spatial distribution of extreme rainfall. Groups of three pairwise associated stations were also obtained and an adjustment was made to a family of extreme value copulas involving the Marshall–Olkin family, whose parameters can be written as a function of Kendall’s τ association measures of the obtained pairs.

Anaerobic threshold estimation by statistical modelling.

Anaerobic threshold (AT) is usually estimated as a change point problem by visual analysis of the cardiorespiratory response to incremental dynamic exercise. In this study, two phase linear (TPL) models of the linear-linear and linear-quadratic type were used for the estimation of AT. The correlation coefficient between the classical and statistical approaches was 0.88, and 0.89 after outlier exclusion. The TPL models provide a simple method for estimating AT that can be easily implemented using a digital computer for the automatic pattern recognition of AT.

Statistical modelling of variability in sediment-water nutrient and oxygen fluxes

Peer reviewed

Finite mixture regression model with random effects: application to neonatal hospital length of stay

A two-component mixture regression model that allows simultaneously for heterogeneity and dependency among observations is proposed. By specifying random effects explicitly in the linear predictor of the mixture probability and the mixture components, parameter estimation is achieved by maximising the corresponding best linear unbiased prediction type log-likelihood. Approximate residual maximum likelihood estimates are obtained via an EM algorithm in the manner of generalised linear mixed model (GLMM). The method can be extended to a g-component mixture regression model with the component density from the exponential family, leading to the development of the class of finite mixture GLMM. For illustration, the method is applied to analyse neonatal length of stay (LOS). It is shown that identification of pertinent factors that influence hospital LOS can provide important information for health care planning and resource allocation. (C) 2002 Elsevier Science B.V. All rights reserved.

Response-based segmentation using Finite Mixture Partial Least Squares: theoretical foundations and an application to American Customer Satisfaction Index Data

When applying multivariate analysis techniques in information systems and social science disciplines, such as management information systems (MIS) and marketing, the assumption that the empirical data originate from a single homogeneous population is often unrealistic. When applying a causal modeling approach, such as partial least squares (PLS) path modeling, segmentation is a key issue in coping with the problem of heterogeneity in estimated cause-and-effect relationships. This chapter presents a new PLS path modeling approach which classifies units on the basis of the heterogeneity of the estimates in the inner model. If unobserved heterogeneity significantly affects the estimated path model relationships on the aggregate data level, the methodology will allow homogenous groups of observations to be created that exhibit distinctive path model estimates. The approach will, thus, provide differentiated analytical outcomes that permit more precise interpretations of each segment formed. An application on a large data set in an example of the American customer satisfaction index (ACSI) substantiates the methodology’s effectiveness in evaluating PLS path modeling results.