Predicting future risk of asthma exacerbations using individual conditional probabilities

Determination of future risk of exacerbations is a key issue in the management of asthma. We previously developed a method to calculate conditional probabilities (π) of future decreases in lung function by using the daily fluctuations in peak expiratory flow (PEF).

Estimating conditional probability densities for periodic variables

Most of the common techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce three novel techniques for tackling such problems, and investigate their performance using synthetic data. We then apply these techniques to the problem of extracting the distribution of wind vector directions from radar scatterometer data gathered by a remote-sensing satellite.

Estimating conditional volatility with neural networks

It is well known that one of the obstacles to effective forecasting of exchange rates is heteroscedasticity (non-stationary conditional variance). The autoregressive conditional heteroscedastic (ARCH) model and its variants have been used to estimate a time dependent variance for many financial time series. However, such models are essentially linear in form and we can ask whether a non-linear model for variance can improve results just as non-linear models (such as neural networks) for the mean have done. In this paper we consider two neural network models for variance estimation. Mixture Density Networks (Bishop 1994, Nix and Weigend 1994) combine a Multi-Layer Perceptron (MLP) and a mixture model to estimate the conditional data density. They are trained using a maximum likelihood approach. However, it is known that maximum likelihood estimates are biased and lead to a systematic under-estimate of variance. More recently, a Bayesian approach to parameter estimation has been developed (Bishop and Qazaz 1996) that shows promise in removing the maximum likelihood bias. However, up to now, this model has not been used for time series prediction. Here we compare these algorithms with two other models to provide benchmark results: a linear model (from the ARIMA family), and a conventional neural network trained with a sum-of-squares error function (which estimates the conditional mean of the time series with a constant variance noise model). This comparison is carried out on daily exchange rate data for five currencies.

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.

Measuring Avoidable Health Inequality with Realization of Conditional Potential Life Years (RCPLY)

In a series of papers (Tang, Chin and Rao, 2008; and Tang, Petrie and Rao 2006 & 2007), we have tried to improve on a mortality-based health status indicator, namely age-at-death (AAD), and its associated health inequality indicators that measure the distribution of AAD. The main contribution of these papers is to propose a frontier method to separate avoidable and unavoidable mortality risks. This has facilitated the development of a new indicator of health status, namely the Realization of Potential Life Years (RePLY). The RePLY measure is based on the concept of a “frontier country” that, by construction, has the lowest mortality risks for each age-sex group amongst all countries. The mortality rates of the frontier country are used as a proxy for the unavoidable mortality rates, and the residual between the observed mortality rates and the unavoidable mortality rates are considered as avoidable morality rates. In this approach, however, countries at different levels of development are benchmarked against the same frontier country without considering their heterogeneity. The main objective of the current paper is to control for national resources in estimating (conditional) unavoidable and avoidable mortality risks for individual countries. This allows us to construct a new indicator of health status – Realization of Conditional Potential Life Years (RCPLY). The paper presents empirical results from a dataset of life tables for 167 countries from the year 2000, compiled and updated by the World Health Organization. Measures of national average health status and health inequality based on RePLY and RCPLY are presented and compared.

Modelling conditional probability distributions for periodic variables

Most of the common techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce two novel techniques for tackling such problems, and investigate their performance using synthetic data. We then apply these techniques to the problem of extracting the distribution of wind vector directions from radar scatterometer data gathered by a remote-sensing satellite.

Modelling conditional probability distributions for periodic variables

Most of the common techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we apply two novel techniques to the problem of extracting the distribution of wind vector directions from radar catterometer data gathered by a remote-sensing satellite.

Modelling conditional probability distributions for periodic variables

Most conventional techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce three related techniques for tackling such problems, and investigate their performance using synthetic data. We then apply these techniques to the problem of extracting the distribution of wind vector directions from radar scatterometer data gathered by a remote-sensing satellite.

An oracle approach for interaction neighborhood estimation in random fields

We consider the problem of interaction neighborhood estimation from the partial observation of a finite number of realizations of a random field. We introduce a model selection rule to choose estimators of conditional probabilities among natural candidates. Our main result is an oracle inequality satisfied by the resulting estimator. We use then this selection rule in a two-step procedure to evaluate the interacting neighborhoods. The selection rule selects a small prior set of possible interacting points and a cutting step remove from this prior set the irrelevant points. We also prove that the Ising models satisfy the assumptions of the main theorems, without restrictions on the temperature, on the structure of the interacting graph or on the range of the interactions. It provides therefore a large class of applications for our results. We give a computationally efficient procedure in these models. We finally show the practical efficiency of our approach in a simulation study.

Quantifying consistency of Event Related Potentials across subjects based on single-trial topographic analysis

Introduction: Non-invasive brain imaging techniques often contrast experimental conditions across a cohort of participants, obfuscating distinctions in individual performance and brain mechanisms that are better characterised by the inter-trial variability. To overcome such limitations, we developed topographic analysis methods for single-trial EEG data [1]. So far this was typically based on time-frequency analysis of single-electrode data or single independent components. The method's efficacy is demonstrated for event-related responses to environmental sounds, hitherto studied at an average event-related potential (ERP) level. Methods: Nine healthy subjects participated to the experiment. Auditory meaningful sounds of common objects were used for a target detection task [2]. On each block, subjects were asked to discriminate target sounds, which were living or man-made auditory objects. Continuous 64-channel EEG was acquired during the task. Two datasets were considered for each subject including single-trial of the two conditions, living and man-made. The analysis comprised two steps. In the first part, a mixture of Gaussians analysis [3] provided representative topographies for each subject. In the second step, conditional probabilities for each Gaussian provided statistical inference on the structure of these topographies across trials, time, and experimental conditions. Similar analysis was conducted at group-level. Results: Results show that the occurrence of each map is structured in time and consistent across trials both at the single-subject and at group level. Conducting separate analyses of ERPs at single-subject and group levels, we could quantify the consistency of identified topographies and their time course of activation within and across participants as well as experimental conditions. A general agreement was found with previous analysis at average ERP level. Conclusions: This novel approach to single-trial analysis promises to have impact on several domains. In clinical research, it gives the possibility to statistically evaluate single-subject data, an essential tool for analysing patients with specific deficits and impairments and their deviation from normative standards. In cognitive neuroscience, it provides a novel tool for understanding behaviour and brain activity interdependencies at both single-subject and at group levels. In basic neurophysiology, it provides a new representation of ERPs and promises to cast light on the mechanisms of its generation and inter-individual variability.

Counsellor behaviours and patient language during brief motivational interventions: a sequential analysis of speech

AIMS: To investigate empirically the hypothesized relationship between counsellor motivational interviewing (MI) skills and patient change talk (CT) by analysing the articulation between counsellor behaviours and patient language during brief motivational interventions (BMI) addressing at-risk alcohol consumption. DESIGN: Sequential analysis of psycholinguistic codes obtained by two independent raters using the Motivational Interviewing Skill Code (MISC), version 2.0. SETTING: Secondary analysis of data from a randomized controlled trial evaluating the effectiveness of BMI in an emergency department. PARTICIPANTS: A total of 97 patients tape-recorded when receiving BMI. MEASUREMENTS: MISC variables were categorized into three counsellor behaviours (MI-consistent, MI-inconsistent and 'other') and three kinds of patient language (CT, counter-CT (CCT) and utterances not linked with the alcohol topic). Observed transition frequencies, conditional probabilities and significance levels based on odds ratios were computed using sequential analysis software. FINDINGS: MI-consistent behaviours were the only counsellor behaviours that were significantly more likely to be followed by patient CT. Those behaviours were significantly more likely to be followed by patient change exploration (CT and CCT) while MI-inconsistent behaviours and 'other' counsellor behaviours were significantly more likely to be followed by utterances not linked with the alcohol topic and significantly less likely to be followed by CT. MI-consistent behaviours were more likely after change exploration, whereas 'other' counsellor behaviours were more likely only after utterances not linked with the alcohol topic. CONCLUSIONS: Findings lend support to the hypothesized relationship between MI-consistent behaviours and CT, highlight the importance of patient influence on counsellor behaviour and emphasize the usefulness of MI techniques and spirit during brief interventions targeting change enhancement.

The quality of the mother-child relationship during collaborative problem solving /

The quality of the mother-child relationship was examined in relation to joint planning, maternal teaching strategies, maternal emotional support, mutual positive affect and attachment security. Fifty-five grade five children and their mothers participated in a laboratory session comprised of various activities and completed questionnaires to evaluate attachment security. Joint planning and social problem solving were assessed observationally during an origami task. Problem solving effectiveness was unrelated to maternal teaching strategies, maternal encouragement and mutual positive affect. A marginally significant relationship was found between maternal encouragement and active child participation. Attachment security was found to be significantly related to sharing of responsibility during local planning, but only for child autonomous performance. An examination of conditional probabilities revealed that mutual positive affect did not increase the likelihood of subsequent mother-child dyadic regulation. However, mutual positive affect was found to be significantly related to both active child participation and dyadic regulation. The hypothesis predicting a mediational model was not supported. The implications of these findings in the theoretical and empirical literature were considered and suggestions for future research were made.

A note on the accuracy of PAC-likelihood inference with microsatellite data

Stephens and Donnelly have introduced a simple yet powerful importance sampling scheme for computing the likelihood in population genetic models. Fundamental to the method is an approximation to the conditional probability of the allelic type of an additional gene, given those currently in the sample. As noted by Li and Stephens, the product of these conditional probabilities for a sequence of draws that gives the frequency of allelic types in a sample is an approximation to the likelihood, and can be used directly in inference. The aim of this note is to demonstrate the high level of accuracy of "product of approximate conditionals" (PAC) likelihood when used with microsatellite data. Results obtained on simulated microsatellite data show that this strategy leads to a negligible bias over a wide range of the scaled mutation parameter theta. Furthermore, the sampling variance of likelihood estimates as well as the computation time are lower than that obtained with importance sampling on the whole range of theta. It follows that this approach represents an efficient substitute to IS algorithms in computer intensive (e.g. MCMC) inference methods in population genetics. (c) 2006 Elsevier Inc. All rights reserved.

Essays on incentives, procurement and regulation

Esta tese é composta de três artigos. No primeiro artigo, "Simple Contracts under Simultaneous Adverse Selection and Moral Hazard", é considerado um problema de principal-agente sob a presença simultânea dos problemas de risco moral e seleção adversa, em que a dimensão de seleção adversa se dá sobre as distribuições de probabilidade condicionais as ações do agente. No segundo artigo, "Public-Private Partnerships in the Presence of Adverse Selection" é analisada a otimalidade de parcerias público-privadas sob a presença de seleção adversa. No terceiro artigo, "Regulation Under Stock Market Information Disclosure", por sua vez, é considerado o problema da regulação de firmas de capital aberto, onde as firmas possuem incentivos para mandar sinais opostos para o regulador e o mercado.

Analisi critica di modelli previsionali per le frane in Emilia Romagna

Questa tesi di dottorato è inserita nell’ambito della convenzione tra ARPA_SIMC (che è l’Ente finanziatore), l’Agenzia Regionale di Protezione Civile ed il Dipartimento di Scienze della Terra e Geologico - Ambientali dell’Ateneo di Bologna. L’obiettivo principale è la determinazione di possibili soglie pluviometriche di innesco per i fenomeni franosi in Emilia Romagna che possano essere utilizzate come strumento di supporto previsionale in sala operativa di Protezione Civile. In un contesto geologico così complesso, un approccio empirico tradizionale non è sufficiente per discriminare in modo univoco tra eventi meteo innescanti e non, ed in generale la distribuzione dei dati appare troppo dispersa per poter tracciare una soglia statisticamente significativa. È stato quindi deciso di applicare il rigoroso approccio statistico Bayesiano, innovativo poiché calcola la probabilità di frana dato un certo evento di pioggia (P(A|B)) , considerando non solo le precipitazioni innescanti frane (quindi la probabilità condizionata di avere un certo evento di precipitazione data l’occorrenza di frana, P(B|A)), ma anche le precipitazioni non innescanti (quindi la probabilità a priori di un evento di pioggia, P(A)). L’approccio Bayesiano è stato applicato all’intervallo temporale compreso tra il 1939 ed il 2009. Le isolinee di probabilità ottenute minimizzano i falsi allarmi e sono facilmente implementabili in un sistema di allertamento regionale, ma possono presentare limiti previsionali per fenomeni non rappresentati nel dataset storico o che avvengono in condizioni anomale. Ne sono esempio le frane superficiali con evoluzione in debris flows, estremamente rare negli ultimi 70 anni, ma con frequenza recentemente in aumento. Si è cercato di affrontare questo problema testando la variabilità previsionale di alcuni modelli fisicamente basati appositamente sviluppati a questo scopo, tra cui X – SLIP (Montrasio et al., 1998), SHALSTAB (SHALlow STABility model, Montgomery & Dietrich, 1994), Iverson (2000), TRIGRS 1.0 (Baum et al., 2002), TRIGRS 2.0 (Baum et al., 2008).