Search results for " statistics"
showing 10 items of 1891 documents
Automated Extraction of Motivic Patterns and Application to the Analysis of Debussy’s Syrinx
2009
A methodology for automated extraction of repeated patterns in discrete time series data is presented, dedicated to the discovery of musical motives in symbolic music representations. The basic principle of the approach consists in a search for closed patterns in a multi-dimensional parametric space, comprising various features related to melodic and rhythmic aspects, which can be organized into note-based and interval-based descriptions. The pattern description is further reduced through a lossless pruning of the sequence description. This requires in particular a detailed estimation of the specificity relations between patterns. For instance, a pattern is more specific than its suffix, an…
Meta model of e-learning materials development
2014
The multitude of software tools is available for the creation of learning resources. However the majority of these tools provided by different software producers do not have a unified mechanism by means of which it would be possible to search and reuse the existing learning resources or their elements. To solve this problem the structures of descriptive data can be used. The aim of this paper is to describe a meta-model of e-learning objects and e-learning formats that could be used in the creation of e-learning materials compatible with various e-learning standards. The meta-data models that are used in widely-known learning resourcesâ repositories and their structureâs metadata standa…
A penalized approach to the bivariate logistic regression model for the association between ordinal responses
2014
Bivariate ordered logistic models (BOLMs) are appealing to jointly model the marginal distribution of two ordered responses and their association, given a set of covariates. When the number of categories of the responses increases, the number of global odds ratios (or their re-parametrizations) to be estimated also increases and estimating the association structure becomes crucial for this type of data. In fact, such data could be too "rich" to be fully modelled with an ordinary BOLM while, sometimes, the well-known Dale's model could be too parsimonious to provide a good fit. In addition, when the cross-tabulation of the responses contains some zeros, for a number of model configurations, …
Optimal design of observational studies: overview and synthesis
2016
We review typical design problems encountered in the planning of observational studies and propose a unifying framework that allows us to use the same concepts and notation for different problems. In the framework, the design is defined as a probability measure in the space of observational processes that determine whether the value of a variable is observed for a specific unit at the given time. The optimal design is then defined, according to Bayesian decision theory, to be the one that maximizes the expected utility related to the design. We present examples on the use of the framework and discuss methods for deriving optimal or approximately optimal designs.
Large-Sample Properties of Blind Estimation of the Linear Discriminant Using Projection Pursuit
2021
We study the estimation of the linear discriminant with projection pursuit, a method that is blind in the sense that it does not use the class labels in the estimation. Our viewpoint is asymptotic and, as our main contribution, we derive central limit theorems for estimators based on three different projection indices, skewness, kurtosis and their convex combination. The results show that in each case the limiting covariance matrix is proportional to that of linear discriminant analysis (LDA), an unblind estimator of the discriminant. An extensive comparative study between the asymptotic variances reveals that projection pursuit is able to achieve efficiency equal to LDA when the groups are…
Using complex surveys to estimate the $L_1$-median of a functional variable: application to electricity load curves
2012
Mean profiles are widely used as indicators of the electricity consumption habits of customers. Currently, in \'Electricit\'e De France (EDF), class load profiles are estimated using point-wise mean function. Unfortunately, it is well known that the mean is highly sensitive to the presence of outliers, such as one or more consumers with unusually high-levels of consumption. In this paper, we propose an alternative to the mean profile: the $L_1$-median profile which is more robust. When dealing with large datasets of functional data (load curves for example), survey sampling approaches are useful for estimating the median profile avoiding storing the whole data. We propose here estimators of…
The Geometry of Uniqueness, Sparsity and Clustering in Penalized Estimation
2021
We provide a necessary and sufficient condition for the uniqueness of penalized least-squares estimators whose penalty term is given by a norm with a polytope unit ball, covering a wide range of methods including SLOPE, PACS, fused, clustered and classical LASSO as well as the related method of basis pursuit. We consider a strong type of uniqueness that is relevant for statistical problems. The uniqueness condition is geometric and involves how the row span of the design matrix intersects the faces of the dual norm unit ball, which for SLOPE is given by the signed permutahedron. Further considerations based this condition also allow to derive results on sparsity and clustering features. In …
THE ROLE OF SUNK COSTS IN THE DECISION TO INVEST IN R&D
2009
We present a dynamic empirical model of a firm's R&D decisions that is consistent with the existence of sunk R&D costs, taking into account that these costs may differ between small and large firms, and among different technological regimes. We estimate a multivariate dynamic discrete choice model using firm-level data of Spanish manufacturing for 1990–2000. Conditional on firm heterogeneity and serially correlated unobservable factors, we find that R&D history matters. This true state dependence allows inferring the existence of sunk R&D costs associated with performing R&D. Sunk R&D costs are found to be higher for large, high-tech firms.
Cost and profit efficiency in the Spanish banking sector (1985–1996): a non-parametric approach
2003
The aim of this article is to analyse the efficiency in costs and in profits of the Spanish banking sector (SBS) in the period 1985–1996 using a non-parametric approach. The results obtained show the existence of profit efficiency levels well below those corresponding to cost efficiency, alternative profit efficiency being below standard profit efficiency. These results imply the existence of market power in the setting of prices and/or the existence of differences in the quality of bank output reflected in the differences in prices. With regard to the immediate future, of full economic and monetary integration, the reduction of profit levels associated with higher competitive pressure may …
Self-Localization of Distributed Microphone Arrays Using Directional Statistics with DoA Estimation Reliability
2019
This paper addresses the problem of self-localization of distributed microphone arrays from microphone recordings by following a two-step optimization procedure. In the first step, the relative geometry of the sources and arrays is inferred by the proposed maximum likelihood estimator. It is derived under the assumption that the acquired unit-norm vectors pointing towards the unknown source positions follow a von Mises-Fisher distribution in a D-dimensional space. In the second step, the absolute positions and synchronization offsets between the arrays are estimated from the inferred relative geometry by using the Least Squares procedure. To improve the accuracy of the method, we propose as…