Convex linear combination processes for compositions

Aitchison and Bacon-Shone (1999) considered convex linear combinations ofcompositions. In other words, they investigated compositions of compositions, wherethe mixing composition follows a logistic Normal distribution (or a perturbationprocess) and the compositions being mixed follow a logistic Normal distribution. Inthis paper, I investigate the extension to situations where the mixing compositionvaries with a number of dimensions. Examples would be where the mixingproportions vary with time or distance or a combination of the two. Practicalsituations include a river where the mixing proportions vary along the river, or acrossa lake and possibly with a time trend. This is illustrated with a dataset similar to thatused in the Aitchison and Bacon-Shone paper, which looked at how pollution in aloch depended on the pollution in the three rivers that feed the loch. Here, I explicitlymodel the variation in the linear combination across the loch, assuming that the meanof the logistic Normal distribution depends on the river flows and relative distancefrom the source origins

To select or to weigh: a comparative study of linear combination schemes for superparent-one-dependence estimators

We conduct a large-scale comparative study on linearly combining superparent-one-dependence estimators (SPODEs), a popular family of seminaive Bayesian classifiers. Altogether, 16 model selection and weighing schemes, 58 benchmark data sets, and various statistical tests are employed. This paper's main contributions are threefold. First, it formally presents each scheme's definition, rationale, and time complexity and hence can serve as a comprehensive reference for researchers interested in ensemble learning. Second, it offers bias-variance analysis for each scheme's classification error performance. Third, it identifies effective schemes that meet various needs in practice. This leads to accurate and fast classification algorithms which have an immediate and significant impact on real-world applications. Another important feature of our study is using a variety of statistical tests to evaluate multiple learning methods across multiple data sets.

System identification using a linear combination of cumulant slices

In this paper we develop a new linear approach to identify the parameters of a moving average (MA) model from the statistics of the output. First, we show that, under some constraints, the impulse response of the system can be expressed as a linear combination of cumulant slices. Then, thisresult is used to obtain a new well-conditioned linear methodto estimate the MA parameters of a non-Gaussian process. Theproposed method presents several important differences withexisting linear approaches. The linear combination of slices usedto compute the MA parameters can be constructed from dif-ferent sets of cumulants of different orders, providing a generalframework where all the statistics can be combined. Further-more, it is not necessary to use second-order statistics (the autocorrelation slice), and therefore the proposed algorithm stillprovides consistent estimates in the presence of colored Gaussian noise. Another advantage of the method is that while mostlinear methods developed so far give totally erroneous estimates if the order is overestimated, the proposed approach doesnot require a previous estimation of the filter order. The simulation results confirm the good numerical conditioning of thealgorithm and the improvement in performance with respect to existing methods.

Convex linear combination processes for compositions

Aitchison and Bacon-Shone (1999) considered convex linear combinations of compositions. In other words, they investigated compositions of compositions, where the mixing composition follows a logistic Normal distribution (or a perturbation process) and the compositions being mixed follow a logistic Normal distribution. In this paper, I investigate the extension to situations where the mixing composition varies with a number of dimensions. Examples would be where the mixing proportions vary with time or distance or a combination of the two. Practical situations include a river where the mixing proportions vary along the river, or across a lake and possibly with a time trend. This is illustrated with a dataset similar to that used in the Aitchison and Bacon-Shone paper, which looked at how pollution in a loch depended on the pollution in the three rivers that feed the loch. Here, I explicitly model the variation in the linear combination across the loch, assuming that the mean of the logistic Normal distribution depends on the river flows and relative distance from the source origins

Two-dimensional linear-combination model fitting of magnetic resonance spectra to define the macromolecule baseline using FiTAID, a Fitting Tool for Arrays of Interrelated Datasets

To propose the determination of the macromolecular baseline (MMBL) in clinical 1H MR spectra based on T(1) and T(2) differentiation using 2D fitting in FiTAID, a general Fitting Tool for Arrays of Interrelated Datasets.

Quantitative 1H-magnetic resonance spectroscopy of human brain: Influence of composition and parameterization of the basis set in linear combination model-fitting

Localized short-echo-time (1)H-MR spectra of human brain contain contributions of many low-molecular-weight metabolites and baseline contributions of macromolecules. Two approaches to model such spectra are compared and the data acquisition sequence, optimized for reproducibility, is presented. Modeling relies on prior knowledge constraints and linear combination of metabolite spectra. Investigated was what can be gained by basis parameterization, i.e., description of basis spectra as sums of parametric lineshapes. Effects of basis composition and addition of experimentally measured macromolecular baselines were investigated also. Both fitting methods yielded quantitatively similar values, model deviations, error estimates, and reproducibility in the evaluation of 64 spectra of human gray and white matter from 40 subjects. Major advantages of parameterized basis functions are the possibilities to evaluate fitting parameters separately, to treat subgroup spectra as independent moieties, and to incorporate deviations from straightforward metabolite models. It was found that most of the 22 basis metabolites used may provide meaningful data when comparing patient cohorts. In individual spectra, sums of closely related metabolites are often more meaningful. Inclusion of a macromolecular basis component leads to relatively small, but significantly different tissue content for most metabolites. It provides a means to quantitate baseline contributions that may contain crucial clinical information.

Discrete-time mean variance optimal control of linear systems with Markovian jumps and multiplicative noise

In this article, we consider the stochastic optimal control problem of discrete-time linear systems subject to Markov jumps and multiplicative noise under three kinds of performance criterions related to the final value of the expectation and variance of the output. In the first problem it is desired to minimise the final variance of the output subject to a restriction on its final expectation, in the second one it is desired to maximise the final expectation of the output subject to a restriction on its final variance, and in the third one it is considered a performance criterion composed by a linear combination of the final variance and expectation of the output of the system. We present explicit sufficient conditions for the existence of an optimal control strategy for these problems, generalising previous results in the literature. We conclude this article presenting a numerical example of an asset liabilities management model for pension funds with regime switching.

Model-Based Matching by Linear Combinations of Prototypes

We describe a method for modeling object classes (such as faces) using 2D example images and an algorithm for matching a model to a novel image. The object class models are "learned'' from example images that we call prototypes. In addition to the images, the pixelwise correspondences between a reference prototype and each of the other prototypes must also be provided. Thus a model consists of a linear combination of prototypical shapes and textures. A stochastic gradient descent algorithm is used to match a model to a novel image by minimizing the error between the model and the novel image. Example models are shown as well as example matches to novel images. The robustness of the matching algorithm is also evaluated. The technique can be used for a number of applications including the computation of correspondence between novel images of a certain known class, object recognition, image synthesis and image compression.

Model-Based Matching of Line Drawings by Linear Combinations of Prototypes

We describe a technique for finding pixelwise correspondences between two images by using models of objects of the same class to guide the search. The object models are 'learned' from example images (also called prototypes) of an object class. The models consist of a linear combination ofsprototypes. The flow fields giving pixelwise correspondences between a base prototype and each of the other prototypes must be given. A novel image of an object of the same class is matched to a model by minimizing an error between the novel image and the current guess for the closest modelsimage. Currently, the algorithm applies to line drawings of objects. An extension to real grey level images is discussed.

A constrained recursive least squares algorithm for adaptive combination of multiple models

In this paper, we develop a novel constrained recursive least squares algorithm for adaptively combining a set of given multiple models. With data available in an online fashion, the linear combination coefficients of submodels are adapted via the proposed algorithm.We propose to minimize the mean square error with a forgetting factor, and apply the sum to one constraint to the combination parameters. Moreover an l1-norm constraint to the combination parameters is also applied with the aim to achieve sparsity of multiple models so that only a subset of models may be selected into the final model. Then a weighted l2-norm is applied as an approximation to the l1-norm term. As such at each time step, a closed solution of the model combination parameters is available. The contribution of this paper is to derive the proposed constrained recursive least squares algorithm that is computational efficient by exploiting matrix theory. The effectiveness of the approach has been demonstrated using both simulated and real time series examples.

Combinação linear ponderada na definição de áreas prioritárias à conectividade entre fragmentos florestais em ambiente SIG

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

On linear combinations of L-orthogonal polynomials associated with distributions belonging to symmetric classes

This paper deals with the classes S-3(omega, beta, b) of strong distribution functions defined on the interval [beta(2)/b, b], 0 < beta < b <= infinity, where 2 omega epsilon Z. The classification is such that the distribution function psi epsilon S-3(omega, beta, b) has a (reciprocal) symmetry, depending on omega, about the point beta. We consider properties of the L-orthogonal polynomials associated with psi epsilon S-3(omega, beta, b). Through linear combination of these polynomials we relate them to the L-orthogonal polynomials associated with some omega epsilon S-3(1/2, beta, b). (c) 2004 Elsevier B.V. All rights reserved.

Restauração florestal visando à conservação de recursos hídricos por meio da combinação linear ponderada

This study evaluated using of Multicriteria Evaluation in a GIS, specifically by Weighted Linear Combination Method for generation of map of priority areas for forest restoration in the initial part of River Pardo Basin, SP, in order to water resources conservation. Aiming to define criteria and restrictions it was used Participatory Techniques, and the following factors had been selected: proximity of the hydrographic network, proximity of forest cover, slope and erodibility of soil. To calculate the weight to each factor it was used the decision-making process, known as Analytic Hierarchy Analysis, this method consists of a paired comparison of factors to determine the relative importance of each. According to Weighted Linear Combination, the very high priority areas have a more limited spatial distribution, with an apparent concentration around the water bodies, outlining a buffer to the river system. The proximity factor of the hydrographic network, and enables the connection forestry, contributed, along with the factor of proximity to forest cover, so there would be the definition of most of the areas with the highest priority in the basin, which concentrate the largest areas of forest and native riparian areas along the hydrographic.

Adequação territorial com abordagem multicriterial pela análise da combinação linear ponderada

From the characterization of biophysical attributes of the watershed (slope, soil types, capacity to land use and land cover), this article, used the multi-criteria analysis method – Weighted Linear Combination, defined priority areas for adaptation to the use of land as to its capacity of use. With this methodological approach, were created for the watershed under study, four classes, formed by different combinations of biophysical attributes (discrete data), representing levels of priorities for agricultural land use. The Multicriteria Evaluation in a GIS is suitable for the mapping of priority areas to the suitability of land use in watersheds. The geospatial information on the biophysical environment, generated from the methodological procedures described in this article, has a high positive potential to guide the rational planning of the use of natural resources and territorial occupation, besides serving as a powerful instrument to guide policies and collective processes of decision on the use and land cover.