Reliability measures for local nodes assessment in classification trees

Most of the modem developments with classification trees are aimed at improving their predictive capacity. This article considers a curiously neglected aspect of classification trees, namely the reliability of predictions that come from a given classification tree. In the sense that a node of a tree represents a point in the predictor space in the limit, the aim of this article is the development of localized assessment of the reliability of prediction rules. A classification tree may be used either to provide a probability forecast, where for each node the membership probabilities for each class constitutes the prediction, or a true classification where each new observation is predictively assigned to a unique class. Correspondingly, two types of reliability measure will be derived-namely, prediction reliability and classification reliability. We use bootstrapping methods as the main tool to construct these measures. We also provide a suite of graphical displays by which they may be easily appreciated. In addition to providing some estimate of the reliability of specific forecasts of each type, these measures can also be used to guide future data collection to improve the effectiveness of the tree model. The motivating example we give has a binary response, namely the presence or absence of a species of Eucalypt, Eucalyptus cloeziana, at a given sampling location in response to a suite of environmental covariates, (although the methods are not restricted to binary response data).

Successive nonparametric estimation of conditional distributions

Spatial characterization of non-Gaussian attributes in earth sciences and engineering commonly requires the estimation of their conditional distribution. The indicator and probability kriging approaches of current nonparametric geostatistics provide approximations for estimating conditional distributions. They do not, however, provide results similar to those in the cumbersome implementation of simultaneous cokriging of indicators. This paper presents a new formulation termed successive cokriging of indicators that avoids the classic simultaneous solution and related computational problems, while obtaining equivalent results to the impractical simultaneous solution of cokriging of indicators. A successive minimization of the estimation variance of probability estimates is performed, as additional data are successively included into the estimation process. In addition, the approach leads to an efficient nonparametric simulation algorithm for non-Gaussian random functions based on residual probabilities.

The transform likelihood ratio method for rare event simulation with heavy tails

We present a novel method, called the transform likelihood ratio (TLR) method, for estimation of rare event probabilities with heavy-tailed distributions. Via a simple transformation ( change of variables) technique the TLR method reduces the original rare event probability estimation with heavy tail distributions to an equivalent one with light tail distributions. Once this transformation has been established we estimate the rare event probability via importance sampling, using the classical exponential change of measure or the standard likelihood ratio change of measure. In the latter case the importance sampling distribution is chosen from the same parametric family as the transformed distribution. We estimate the optimal parameter vector of the importance sampling distribution using the cross-entropy method. We prove the polynomial complexity of the TLR method for certain heavy-tailed models and demonstrate numerically its high efficiency for various heavy-tailed models previously thought to be intractable. We also show that the TLR method can be viewed as a universal tool in the sense that not only it provides a unified view for heavy-tailed simulation but also can be efficiently used in simulation with light-tailed distributions. We present extensive simulation results which support the efficiency of the TLR method.

Extinction times for a general birth, death and catastrophe process

The birth, death and catastrophe process is an extension of the birth-death process that incorporates the possibility of reductions in population of arbitrary size. We will consider a general form of this model in which the transition rates are allowed to depend on the current population size in an arbitrary manner. The linear case, where the transition rates are proportional to current population size, has been studied extensively. In particular, extinction probabilities, the expected time to extinction, and the distribution of the population size conditional on nonextinction (the quasi-stationary distribution) have all been evaluated explicitly. However, whilst these characteristics are of interest in the modelling and management of populations, processes with linear rate coefficients represent only a very limited class of models. We address this limitation by allowing for a wider range of catastrophic events. Despite this generalisation, explicit expressions can still be found for the expected extinction times.

General practitioners' role in preventive medicine: scenario analysis using alcohol as a case study

The purpose of this analysis is threefold: first, to extract from the literature, current levels of GP detection of at-risk drinking by their patients, rates at which general practitioners (GPs) offer an intervention; and the effectiveness of these interventions; secondly, to develop a model based on this literature to be used in conjunction with scenario analysis; and thirdly, to consider the cost implications of current efforts and various scenarios. This study deals specifically with Australian general practice. A two-step procedure is used in the scenario analysis, which involves identifying opportunities for detection, intervention, effectiveness and assigning probabilities to outcomes. The results suggest that increasing rates of GP intervention achieves greatest benefit and return on resource use. For every 5% point increase in the rate of GP intervention, an additional 26 754 at-risk drinkers modify their drinking behaviour at a cost of $231.45 per patient. This compares with a cost per patient modifying drinking behaviour of $232.60 and $208.31 for every 5% point increase in the rates of detection and effectiveness, respectively. The knowledge, skill and attitude of practitioners toward drinking are significant, and they can be the prime motivators in persuading their patients to modify drinking behaviour.

Precautionary principles: a jurisdiction-free framework for decision-making under risk

Fundamental principles of precaution are legal maxims that ask for preventive actions, perhaps as contingent interim measures while relevant information about causality and harm remains unavailable, to minimize the societal impact of potentially severe or irreversible outcomes. Such principles do not explain how to make choices or how to identify what is protective when incomplete and inconsistent scientific evidence of causation characterizes the potential hazards. Rather, they entrust lower jurisdictions, such as agencies or authorities, to make current decisions while recognizing that future information can contradict the scientific basis that supported the initial decision. After reviewing and synthesizing national and international legal aspects of precautionary principles, this paper addresses the key question: How can society manage potentially severe, irreversible or serious environmental outcomes when variability, uncertainty, and limited causal knowledge characterize their decision-making? A decision-analytic solution is outlined that focuses on risky decisions and accounts for prior states of information and scientific beliefs that can be updated as subsequent information becomes available. As a practical and established approach to causal reasoning and decision-making under risk, inherent to precautionary decision-making, these (Bayesian) methods help decision-makers and stakeholders because they formally account for probabilistic outcomes, new information, and are consistent and replicable. Rational choice of an action from among various alternatives-defined as a choice that makes preferred consequences more likely-requires accounting for costs, benefits and the change in risks associated with each candidate action. Decisions under any form of the precautionary principle reviewed must account for the contingent nature of scientific information, creating a link to the decision-analytic principle of expected value of information (VOI), to show the relevance of new information, relative to the initial ( and smaller) set of data on which the decision was based. We exemplify this seemingly simple situation using risk management of BSE. As an integral aspect of causal analysis under risk, the methods developed in this paper permit the addition of non-linear, hormetic dose-response models to the current set of regulatory defaults such as the linear, non-threshold models. This increase in the number of defaults is an important improvement because most of the variants of the precautionary principle require cost-benefit balancing. Specifically, increasing the set of causal defaults accounts for beneficial effects at very low doses. We also show and conclude that quantitative risk assessment dominates qualitative risk assessment, supporting the extension of the set of default causal models.

Symmetry contaminations in reactive scattering through long-lived collision complexes

We extend our Lanczos subspace time-independent wave packet method [J. Chem. Phys. 116 (2002) 2354] to investigate the issue of symmetry contaminations for the challenging deep-well H + O-2 reaction. Our central objective is to address the issue of whether significant symmetry contamination can occur if a wavepacket initially possessing the correct O-O exchange symmetry is propagated over tens of thousands of recursive steps using a basis which does not explicitly enforce the correct symmetry, and if so how seriously this affects the results. We find that symmetry contamination does exist where the symmetry constraint is not explicitly enforced in the basis. While it affects individual resonances and the associated peak amplitudes, the overall shape of the more averaged quantities such as total reaction probabilities and vibrational branching ratios are not seriously affected. (C) 2004 Elsevier B.V. All rights reserved.

Exposing extinction risk analysis to pathogens: Is disease just another form of density dependence?

In the United States and several other countries., the development of population viability analyses (PVA) is a legal requirement of any species survival plan developed for threatened and endangered species. Despite the importance of pathogens in natural populations, little attention has been given to host-pathogen dynamics in PVA. To study the effect of infectious pathogens on extinction risk estimates generated from PVA, we review and synthesize the relevance of host-pathogen dynamics in analyses of extinction risk. We then develop a stochastic, density-dependent host-parasite model to investigate the effects of disease on the persistence of endangered populations. We show that this model converges on a Ricker model of density dependence under a suite of limiting assumptions, including. a high probability that epidemics will arrive and occur. Using this modeling framework, we then quantify: (1) dynamic differences between time series generated by disease and Ricker processes with the same parameters; (2) observed probabilities of quasi-extinction for populations exposed to disease or self-limitation; and (3) bias in probabilities of quasi-extinction estimated by density-independent PVAs when populations experience either form of density dependence. Our results suggest two generalities about the relationships among disease, PVA, and the management of endangered species. First, disease more strongly increases variability in host abundance and, thus, the probability of quasi-extinction, than does self-limitation. This result stems from the fact that the effects and the probability of occurrence of disease are both density dependent. Second, estimates of quasi-extinction are more often overly optimistic for populations experiencing disease than for those subject to self-limitation. Thus, although the results of density-independent PVAs may be relatively robust to some particular assumptions about density dependence, they are less robust when endangered populations are known to be susceptible to disease. If potential management actions involve manipulating pathogens, then it may be useful to. model disease explicitly.

Improving the efficiency of wildlife monitoring by estimating detectability: a case study of foxes (Vulpes vulpes) on the Eyre Peninsula, South Australia

Demonstrating the existence of trends in monitoring data is of increasing practical importance to conservation managers wishing to preserve threatened species or reduce the impact of pest species. However, the ability to do so can be compromised if the species in question has low detectability and the true occupancy level or abundance of the species is thus obscured. Zero-inflated models that explicitly model detectability improve the ability to make sound ecological inference in such situations. In this paper we apply an occupancy model including detectability to data from the initial stages of a fox-monitoring program on the Eyre Peninsula, South Australia. We find that detectability is extremely low (< 18%) and varies according to season and the presence or absence of roadside vegetation. We show that simple methods of using monitoring data to inform management, such as plotting the raw data or performing logistic regression, fail to accurately diagnose either the status of the fox population or its trajectory over time. We use the results of the detectability model to consider how future monitoring could be redesigned to achieve efficiency gains. A wide range of monitoring programs could benefit from similar analyses, as part of an active adaptive approach to improving monitoring and management.

Coherent-state linear optical quantum computing gates using simplified diagonal superposition resource states

In this paper we explore the possibility of fundamental tests for coherent-state optical quantum computing gates [ T. C. Ralph et al. Phys. Rev. A 68 042319 (2003)] using sophisticated but not unrealistic quantum states. The major resource required in these gates is a state diagonal to the basis states. We use the recent observation that a squeezed single-photon state [S(r)∣1⟩] approximates well an odd superposition of coherent states (∣α⟩−∣−α⟩) to address the diagonal resource problem. The approximation only holds for relatively small α, and hence these gates cannot be used in a scalable scheme. We explore the effects on fidelities and probabilities in teleportation and a rotated Hadamard gate.

Market failure in long-term private health insurance markets: a proposed solution

Recently, private health insurance rates have declined in many countries. In places requiring community rating in their health insurance premiums, a major cause is age-based adverse selection. However, even in countries without community rating, a de facto type of partial community rating tends to occur. In this note, a modified version of Pauly et al.'s guaranteed renewability model, which addresses the problem of age-based adverse selection (Pauly et al., 1995) is presented. Their model is extended from three to 35 periods. Also, probabilities are allowed to increase by age for low-risk types using actual age-based probabilities. This extension of their work shows that private health insurance contracts available stray far from optimal contracts that deal with age-based adverse selection. This suggests that government actions to affect private insurance options are warranted.

Lifetime cover in private insurance markets

In the last few decades, private health insurance rates have declined in many countries. In countries and states with community rating, a major cause is adverse selection. In order to address age-based adverse selection, Australia has recently begun a novel approach which imposes stiff penalties for buying private insurance later in life, when expected costs are higher. In this paper, we analyze Australiarsquos Lifetime Cover in the context of a modified version of the Rothschild-Stiglitz insurance model (Rothschild and Stiglitz, 1976). We allow empirically-based probabilities to increase by age for low-risk types. The model highlights the shortcomings of the Australian plan. Based on empirically-based probabilities of illness, we predict that Lifetime Cover will not arrest adverse selection. The model has many policy implications for government regulation encouraging long-term health coverage.

Genetic time-series analysis identifies a major QTL for in vivo alcohol metabolism not predicted by in vitro studies of structural protein polymorphism at the ADH1B or ADH1C loci

After ingestion of a standardized dose of ethanol, alcohol concentrations were assessed, over 3.5 hours from blood (six readings) and breath (10 readings) in a sample of 412 MZ and DZ twins who took part in an Alcohol Challenge Twin Study (ACTS). Nearly all participants were subsequently genotyped on two polymorphic SNPs in the ADH1B and ADH1C loci known to affect in vitro ADH activity. In the DZ pairs, 14 microsatellite markers covering a 20.5 cM region on chromosome 4 that includes the ADH gene family were assessed, Variation in the timed series of autocorrelated blood and breath alcohol readings was studied using a bivariate simplex design. The contribution of a quantitative trait locus (QTL) or QTL's linked to the ADH region was estimated via a mixture of likelihoods weighted by identity-by-descent probabilities. The effects of allelic substitution at the ADH1B and ADH1C loci were estimated in the means part of the model simultaneously with the effects sex and age. There was a major contribution to variance in alcohol metabolism due to a QTL which accounted for about 64% of the additive genetic covariation common to both blood and breath alcohol readings at the first time point. No effects of the ADH1B*47His or ADH1C*349Ile alleles on in vivo metabolism were observed, although these have been shown to have major effects in vitro. This implies that there is a major determinant of variation for in vivo alcohol metabolism in the ADH region that is not accounted for by these polymorphisms. Earlier analyses of these data suggested that alcohol metabolism is related to drinking behavior and imply that this QTL may be protective against alcohol dependence.

A cost-effectiveness analysis of fluvastatin in patients with diabetes after successful percutaneous coronary intervention

Background: The Lescol Intervention Prevention Study (LIPS) was a multinational randomized controlled trial that showed a 47% reduction in the relative risk of cardiac death and a 22% reduction in major adverse cardiac events (MACEs) from the routine use of fluvastatin, compared with controls, in patients undergoing percutaneous coronary intervention (PCI, defined as angioplasty with or without stents). In this study, MACEs included cardiac death, nonfatal myocardial infarction, and subsequent PCI and coronary artery bypass graft. Diabetes was the greatest risk factor for MACEs. Objective: This study estimated the cost-effectiveness of fluvastatin when used for secondary prevention of MACEs after PCI in people with diabetes. Methods: A post hoc subgroup analysis of patients with diabetes from the LIPS was used to estimate the effectiveness of fluvastatin in reducing myocardial infarction, revascularization, and cardiac death. A probabilistic Markov model was developed using United Kingdom resource and cost data to estimate the additional costs and quality-adjusted life-years (QALYs) gained over 10 years from the perspective of the British National Health Service. The model contained 6 health states, and the transition probabilities were derived from the LIPS data. Crossover from fluvastatin to other lipid-lowering drugs, withdrawal from fluvastatin, and the use of lipid-lowering drugs in the control group were included. Results: In the subgroup of 202 patients with diabetes in the LIPS trial, 18 (15.0%) of 120 fluvastatin patients and 21 (25.6%) of 82 control participants were insulin dependent (P = NS). Compared with the control group, patients treated with fluvastatin can expect to gain an additional mean (SD) of 0.196 (0.139) QALY per patient over 10 years (P < 0.001) and will cost the health service an additional mean (SD) of 10 (E448) (P = NS) (mean [SD] US $16 [$689]). The additional cost per QALY gained was;(51 (US $78). The key determinants of cost-effectiveness included the probabilities of repeat interventions, cardiac death, the cost of fluvastatin, and the time horizon used for the evaluation. Conclusion: Fluvastatin was an economically efficient treatment to prevent MACEs in these patients with diabetes undergoing PCI.

A review of ACI recommendations for prevention of plastic cracking

The ACI recommendations for the prevention of cracking of plastic concrete attempt to eliminate such cracking by ensuring that the rate of evaporation from unprotected concrete surfaces does not exceed the estimated rate of bleed water production. The current recommendations, however do not account for the large scatter of the underlying experimental evaporation data nor the effect of altitude on evaporation rate. Ignoring the scatter of the evaporation data frequently leads to an unacceptably high probability that the evaporation rate will exceed the bleed rate. Ignoring the effect of altitude leads to similar high probabilities, but in only a comparatively small number of cases. Simple modifications of the ACI recommendations are suggested that can account for both effects. However; insufficient data on the variability of bleed rates are currently available to allow the scatter of the evaporation data to be accounted for completely.