The Ministry of Social Development in Mexico is in charge of creating and assigning social programmes targeting specific needs in the population for the improvement of quality of life. To better target the social programmes, the Ministry is aimed to find clusters of households with the same needs based on demographic characteristics as well as poverty conditions of the household. Available data consists of continuous, ordinal, and nominal variables and the observations are not iid but come from a survey sample based on a complex design. We propose a Bayesian nonparametric mixture model that jointly models a set of latent variables, as in an underlying variable response approach, associated to the observed mixed scale data and accommodates for the different sampling probabilities. The performance of the model is assessed via simulated data. A full analysis of socio-economic conditions in households in the Mexican State of Mexico is presented. the country. To fulfil this objective, SEDESOL creates social programmes to target specific needs in the population.Currently, each existing social programme has its own rules of operation and its own way of selecting the potential households to be benefited by the programme, but in general they all use household income as the main selection criterion. In order to simplify the selection of potential candidates and to better target the programmes to the correct population, SEDESOL wants to create a clustering of households based on needs, socio-economical and demographical features as well as poverty conditions.In 2009, the Council for National Evaluation (CONEVAL, 2009) proposed a methodology for measuring the poverty conditions in households in terms of multiple indicators. These include the income dimension, social deprivations and social cohesion. As a result of this new methodology a multi-dimensional measurement was created based on seven indicators: income, and six deprivation indicators such as education, access to health services, access to social security, housing quality, access to basic public services and access to feeding. These will be the core set of variables used in the clustering later on.In January of 2004 a new general law for social development was passed in Mexico. It establishes that poverty measurements must be calculated every two years at a state level and every five years at a municipality level. For these purposes, the National Institute for Official Statistics (INEGI) implemented a survey based on a complex design of households.This national survey of income and expenses in households (ENIGH) through a module of socio-economic conditions (MCS) collects the required information to produce the multidimensional poverty indicators at a household level and in some cases at an individual level.These household poverty indicators are then expanded with the corresponding sampling design weights to produce poverty indicators at a state level.Model-based clustering approaches (McLachlan and Basford, 1988; Banfiedl and Raftery, 1993) rely on a probability ...

Model-based Bayesian evidence combination leads to models with multiple parameteric modules. In this setting the effects of model misspecification in one of the modules may in some cases be ameliorated by cutting the flow of information from the misspecified module. Semi-Modular Inference (SMI) is a framework allowing partial cuts which modulate but do not completely cut the flow of information between modules. We show that SMI is part of a family of inference procedures which implement partial cuts. It has been shown that additive losses determine an optimal, valid and order-coherent belief update. The losses which arise in Cut models and SMI are not additive. However, like the prequential score function, they have a kind of prequential additivity which we define. We show that prequential additivity is sufficient to determine the optimal valid and order-coherent belief update and that this belief update coincides with the belief update in each of our SMI schemes.

We introduce Neural Contextual Anomaly Detection (NCAD), a framework for anomaly detection on time series that scales seamlessly from the unsupervised to supervised setting, and is applicable to both univariate and multivariate time series. This is achieved by combining recent developments in representation learning for multivariate time series, with techniques for deep anomaly detection originally developed for computer vision that we tailor to the time series setting. Our window-based approach facilitates learning the boundary between normal and anomalous classes by injecting generic synthetic anomalies into the available data. NCAD can effectively take advantage of domain knowledge and of any available training labels. We demonstrate empirically on standard benchmark datasets that our approach obtains a state-of-the-art performance in the supervised, semi-supervised, and unsupervised settings.

We introduce Neural Contextual Anomaly Detection (NCAD), a framework for anomaly detection on time series that scales seamlessly from the unsupervised to supervised setting, and is applicable to both univariate and multivariate time series. This is achieved by effectively combining recent developments in representation learning for multivariate time series, with techniques for deep anomaly detection originally developed for computer vision that we tailor to the time series setting. Our window-based approach facilitates learning the boundary between normal and anomalous classes by injecting generic synthetic anomalies into the available data. Moreover, our method can effectively take advantage of all the available information, be it as domain knowledge, or as training labels in the semi-supervised setting. We demonstrate empirically on standard benchmark datasets that our approach obtains a state-of-the-art performance in these settings. * Equal contribution. † Work done while working at AWS AI Labs.Preprint. Under review.

No abstract