Replication-based inference algorithms for hard computational problems

Inference algorithms based on evolving interactions between replicated solutions are introduced and analyzed on a prototypical NP-hard problem: the capacity of the binary Ising perceptron. The efficiency of the algorithm is examined numerically against that of the parallel tempering algorithm, showing improved performance in terms of the results obtained, computing requirements and simplicity of implementation. © 2013 American Physical Society.

Likelihood-Based Inference for Partially Observed Multi-Type Markov Branching Processes

Thesis (Ph.D.)--University of Washington, 2016-08

Finite Population Inference for Causal Parameters

Thesis (Ph.D.)--University of Washington, 2016-08

Less-is-more effects in knowledge-based heuristic inference

Inference on the basis of recognition alone is assumed to occur prior to accessing further information (Pachur & Hertwig, 2006). A counterintuitive result of this is the “less-is-more” effect: a drop in the accuracy with which choices are made as to which of two or more items scores highest on a given criterion as more items are learned (Frosch, Beaman & McCloy, 2007; Goldstein & Gigerenzer, 2002). In this paper, we show that less-is-more effects are not unique to recognition-based inference but can also be observed with a knowledge-based strategy provided two assumptions, limited information and differential access, are met. The LINDA model which embodies these assumptions is presented. Analysis of the less-is-more effects predicted by LINDA and by recognition-driven inference shows that these occur for similar reasons and casts doubt upon the “special” nature of recognition-based inference. Suggestions are made for empirical tests to compare knowledge-based and recognition-based less-is-more effects

A little learning is a dangerous thing: an experimental demonstration of ignorance-driven inference

Studies of ignorance-driven decision making have been employed to analyse when ignorance should prove advantageous on theoretical grounds or else they have been employed to examine whether human behaviour is consistent with an ignorance-driven inference strategy (e. g., the recognition heuristic). In the current study we examine whether-under conditions where such inferences might be expected-the advantages that theoretical analyses predict are evident in human performance data. A single experiment shows that, when asked to make relative wealth judgements, participants reliably use recognition as a basis for their judgements. Their wealth judgements under these conditions are reliably more accurate when some of the target names are unknown than when participants recognize all of the names (a "less-is-more effect"). These results are consistent across a number of variations: the number of options given to participants and the nature of the wealth judgement. A basic model of recognition-based inference predicts these effects.

When does ignorance make us smart? Additional factors guiding heuristic inference

“Fast & frugal” heuristics represent an appealing way of implementing bounded rationality and decision-making under pressure. The recognition heuristic is the simplest and most fundamental of these heuristics. Simulation and experimental studies have shown that this ignorance-driven heuristic inference can prove superior to knowledge based inference (Borges, Goldstein, Ortman & Gigerenzer, 1999; Goldstein & Gigerenzer, 2002) and have shown how the heuristic could develop from ACT-R’s forgetting function (Schooler & Hertwig, 2005). Mathematical analyses also demonstrate that, under certain conditions, a “less-is-more effect” will always occur (Goldstein & Gigerenzer, 2002). The further analyses presented in this paper show, however, that these conditions may constitute a special case and that the less-is-more effect in decision-making is subject to the moderating influence of the number of options to be considered and the framing of the question.

Improved likelihood inference in beta regression

We consider the issue of performing accurate small-sample likelihood-based inference in beta regression models, which are useful for modelling continuous proportions that are affected by independent variables. We derive small-sample adjustments to the likelihood ratio statistic in this class of models. The adjusted statistics can be easily implemented from standard statistical software. We present Monte Carlo simulations showing that inference based on the adjusted statistics we propose is much more reliable than that based on the usual likelihood ratio statistic. A real data example is presented.

Design-based random permutation models with auxiliary information

We extend the random permutation model to obtain the best linear unbiased estimator of a finite population mean accounting for auxiliary variables under simple random sampling without replacement (SRS) or stratified SRS. The proposed method provides a systematic design-based justification for well-known results involving common estimators derived under minimal assumptions that do not require specification of a functional relationship between the response and the auxiliary variables.

Extended truncated Tweedie-Poisson model

It has been argued that by truncating the sample space of the negative binomial and of the inverse Gaussian-Poisson mixture models at zero, one is allowed to extend the parameter space of the model. Here that is proved to be the case for the more general three parameter Tweedie-Poisson mixture model. It is also proved that the distributions in the extended part of the parameter space are not the zero truncation of mixed poisson distributions and that, other than for the negative binomial, they are not mixtures of zero truncated Poisson distributions either. By extending the parameter space one can improve the fit when the frequency of one is larger and the right tail is heavier than is allowed by the unextended model. Considering the extended model also allows one to use the basic maximum likelihood based inference tools when parameter estimates fall in the extended part of the parameter space, and hence when the m.l.e. does not exist under the unextended model. This extended truncated Tweedie-Poisson model is proved to be useful in the analysis of words and species frequency count data.

Uncertainty Quantification Of A Semi-Supervised Support Vector Regression Reservoir Model

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. Recent advances in machine learning offer a novel approach to model spatial distribution of petrophysical properties in complex reservoirs alternative to geostatistics. The approach is based of semisupervised learning, which handles both ?labelled? observed data and ?unlabelled? data, which have no measured value but describe prior knowledge and other relevant data in forms of manifolds in the input space where the modelled property is continuous. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic geological features and describe stochastic variability and non-uniqueness of spatial properties. On the other hand, it is able to capture and preserve key spatial dependencies such as connectivity of high permeability geo-bodies, which is often difficult in contemporary petroleum reservoir studies. Semi-supervised SVR as a data driven algorithm is designed to integrate various kind of conditioning information and learn dependences from it. The semi-supervised SVR model is able to balance signal/noise levels and control the prior belief in available data. In this work, stochastic semi-supervised SVR geomodel is integrated into Bayesian framework to quantify uncertainty of reservoir production with multiple models fitted to past dynamic observations (production history). Multiple history matched models are obtained using stochastic sampling and/or MCMC-based inference algorithms, which evaluate posterior probability distribution. Uncertainty of the model is described by posterior probability of the model parameters that represent key geological properties: spatial correlation size, continuity strength, smoothness/variability of spatial property distribution. The developed approach is illustrated with a fluvial reservoir case. The resulting probabilistic production forecasts are described by uncertainty envelopes. The paper compares the performance of the models with different combinations of unknown parameters and discusses sensitivity issues.

Semi-Parametric Weak Instrument Regressions with an Application to the Risk-Return Trade-off

Recent work shows that a low correlation between the instruments and the included variables leads to serious inference problems. We extend the local-to-zero analysis of models with weak instruments to models with estimated instruments and regressors and with higher-order dependence between instruments and disturbances. This makes this framework applicable to linear models with expectation variables that are estimated non-parametrically. Two examples of such models are the risk-return trade-off in finance and the impact of inflation uncertainty on real economic activity. Results show that inference based on Lagrange Multiplier (LM) tests is more robust to weak instruments than Wald-based inference. Using LM confidence intervals leads us to conclude that no statistically significant risk premium is present in returns on the S&P 500 index, excess holding yields between 6-month and 3-month Treasury bills, or in yen-dollar spot returns.

Densités de copules archimédiennes hiérarchiques

Les copulas archimédiennes hiérarchiques ont récemment gagné en intérêt puisqu’elles généralisent la famille de copules archimédiennes, car elles introduisent une asymétrie partielle. Des algorithmes d’échantillonnages et des méthodes ont largement été développés pour de telles copules. Néanmoins, concernant l’estimation par maximum de vraisemblance et les tests d’adéquations, il est important d’avoir à disposition la densité de ces variables aléatoires. Ce travail remplie ce manque. Après une courte introduction aux copules et aux copules archimédiennes hiérarchiques, une équation générale sur les dérivées des noeuds et générateurs internes apparaissant dans la densité des copules archimédiennes hiérarchique. sera dérivée. Il en suit une formule tractable pour la densité des copules archimédiennes hiérarchiques. Des exemples incluant les familles archimédiennes usuelles ainsi que leur transformations sont présentés. De plus, une méthode numérique efficiente pour évaluer le logarithme des densités est présentée.

Inférence robuste à la présence des valeurs aberrantes dans les enquêtes

Cette thèse comporte trois articles dont un est publié et deux en préparation. Le sujet central de la thèse porte sur le traitement des valeurs aberrantes représentatives dans deux aspects importants des enquêtes que sont : l’estimation des petits domaines et l’imputation en présence de non-réponse partielle. En ce qui concerne les petits domaines, les estimateurs robustes dans le cadre des modèles au niveau des unités ont été étudiés. Sinha & Rao (2009) proposent une version robuste du meilleur prédicteur linéaire sans biais empirique pour la moyenne des petits domaines. Leur estimateur robuste est de type «plugin», et à la lumière des travaux de Chambers (1986), cet estimateur peut être biaisé dans certaines situations. Chambers et al. (2014) proposent un estimateur corrigé du biais. En outre, un estimateur de l’erreur quadratique moyenne a été associé à ces estimateurs ponctuels. Sinha & Rao (2009) proposent une procédure bootstrap paramétrique pour estimer l’erreur quadratique moyenne. Des méthodes analytiques sont proposées dans Chambers et al. (2014). Cependant, leur validité théorique n’a pas été établie et leurs performances empiriques ne sont pas pleinement satisfaisantes. Ici, nous examinons deux nouvelles approches pour obtenir une version robuste du meilleur prédicteur linéaire sans biais empirique : la première est fondée sur les travaux de Chambers (1986), et la deuxième est basée sur le concept de biais conditionnel comme mesure de l’influence d’une unité de la population. Ces deux classes d’estimateurs robustes des petits domaines incluent également un terme de correction pour le biais. Cependant, ils utilisent tous les deux l’information disponible dans tous les domaines contrairement à celui de Chambers et al. (2014) qui utilise uniquement l’information disponible dans le domaine d’intérêt. Dans certaines situations, un biais non négligeable est possible pour l’estimateur de Sinha & Rao (2009), alors que les estimateurs proposés exhibent un faible biais pour un choix approprié de la fonction d’influence et de la constante de robustesse. Les simulations Monte Carlo sont effectuées, et les comparaisons sont faites entre les estimateurs proposés et ceux de Sinha & Rao (2009) et de Chambers et al. (2014). Les résultats montrent que les estimateurs de Sinha & Rao (2009) et de Chambers et al. (2014) peuvent avoir un biais important, alors que les estimateurs proposés ont une meilleure performance en termes de biais et d’erreur quadratique moyenne. En outre, nous proposons une nouvelle procédure bootstrap pour l’estimation de l’erreur quadratique moyenne des estimateurs robustes des petits domaines. Contrairement aux procédures existantes, nous montrons formellement la validité asymptotique de la méthode bootstrap proposée. Par ailleurs, la méthode proposée est semi-paramétrique, c’est-à-dire, elle n’est pas assujettie à une hypothèse sur les distributions des erreurs ou des effets aléatoires. Ainsi, elle est particulièrement attrayante et plus largement applicable. Nous examinons les performances de notre procédure bootstrap avec les simulations Monte Carlo. Les résultats montrent que notre procédure performe bien et surtout performe mieux que tous les compétiteurs étudiés. Une application de la méthode proposée est illustrée en analysant les données réelles contenant des valeurs aberrantes de Battese, Harter & Fuller (1988). S’agissant de l’imputation en présence de non-réponse partielle, certaines formes d’imputation simple ont été étudiées. L’imputation par la régression déterministe entre les classes, qui inclut l’imputation par le ratio et l’imputation par la moyenne sont souvent utilisées dans les enquêtes. Ces méthodes d’imputation peuvent conduire à des estimateurs imputés biaisés si le modèle d’imputation ou le modèle de non-réponse n’est pas correctement spécifié. Des estimateurs doublement robustes ont été développés dans les années récentes. Ces estimateurs sont sans biais si l’un au moins des modèles d’imputation ou de non-réponse est bien spécifié. Cependant, en présence des valeurs aberrantes, les estimateurs imputés doublement robustes peuvent être très instables. En utilisant le concept de biais conditionnel, nous proposons une version robuste aux valeurs aberrantes de l’estimateur doublement robuste. Les résultats des études par simulations montrent que l’estimateur proposé performe bien pour un choix approprié de la constante de robustesse.

An ontological approach to competency management

Competency management is a very important part of a well-functioning organisation. Unfortunately competency descriptions are not uniformly specified nor defined across borders: National, sectorial or organisational, leading to an opaque competency description market with a multitude of competency frameworks and competency benchmarks. An ontology is a formalised description of a domain, which enables automated reasoning engines to be built which by utilising the interrelations between entities can make “intelligent” choices in different situations within the domain. Introducing formalised competency ontologies automated tools, such as skill gap analysis, training suggestion generation, job search and recruitment, can be developed, which compare and contrast different competency descriptions on the semantic level. The major problem with defining a common formalised ontology for competencies is that there are so many viewpoints of competencies and competency frameworks. Work within the TRACE project has focused on finding common trends within different competency frameworks in order to allow an intermediate competency description to be made, which other frameworks can reference. This research has shown that competencies can be divided up into “knowledge”, “skills” and what we call “others”. An ontology has been created based on this with a simple structure of different “kinds” of “knowledges” and “skills” using semantic interrelations to define the basic semantic structure of the ontology. A prototype tool for analysing a skill gap analysis has been developed. Personal profiles can be produced using the tool and a skill gap analysis is performed on a desired competency profile by using an ontologically based inference engine, which is able to list closest fit and possible proficiency gaps

Less-is-more effects without the recognition heuristic

Inferences consistent with “recognition-based” decision-making may be drawn for various reasons other than recognition alone. We demonstrate that, for 2-alternative forced-choice decision tasks, less-is-more effects (reduced performance with additional learning) are not restricted to recognition-based inference but can also be seen in circumstances where inference is knowledge-based but item knowledge is limited. One reason why such effects may not be observed more widely is the dependence of the effect on specific values for the validity of recognition and knowledge cues. We show that both recognition and knowledge validity may vary as a function of the number of items recognized. The implications of these findings for the special nature of recognition information, and for the investigation of recognition-based inference, are discussed