Examinando por Materia "principal component analysis"
Mostrando 1 - 4 de 4
Resultados por página
Opciones de ordenación
Ítem Directional multivariate extremes in environmental phenomena(John Wiley and Sons Ltd, 2017-03-01) Torres, R.; De michele, C.; Laniado, H.; Lillo, R.E.; Universidad EAFIT. Escuela de Ciencias; Modelado MatemáticoSeveral environmental phenomena can be described by different correlated variables that must be considered jointly in order to be more representative of the nature of these phenomena. For such events, identification of extremes is inappropriate if it is based on marginal analysis. Extremes have usually been linked to the notion of quantile, which is an important tool to analyze risk in the univariate setting. We propose to identify multivariate extremes and analyze environmental phenomena in terms of the directional multivariate quantile, which allows us to analyze the data considering all the variables implied in the phenomena, as well as look at the data in interesting directions that can better describe an environmental catastrophe. Because there are many references in the literature that propose extremes detection based on copula models, we also generalize the copula method by introducing the directional approach. Advantages and disadvantages of the nonparametric proposal that we introduce and the copula methods are provided in the paper. We show with simulated and real data sets how by considering the first principal component direction we can improve the visualization of extremes. Finally, two cases of study are analyzed: a synthetic case of flood risk at a dam (a three-variable case) and a real case study of sea storms (a five-variable case). Copyright © 2017 John Wiley & Sons, Ltd.Ítem Principal component analysis -PCA- and delone triangulations for PL approximation C-1-continuous 1-manifolds in R-N(ACTA PRESS, 2004-01-01) Ruiz, OE; Cadavid, CA; Garcia, MJ; Martinod, R; Universidad EAFIT. Departamento de Ingeniería Mecánica; Laboratorio CAD/CAM/CAEA Method is presented which combines statistical (Principal Component Analysis) and deterministic (Voronoi-Delone) methods to find Piecewise Linear approximations of curves C-i(u) in R-3 sampled with statistical noise. If the curves are self-intersecting, there are a finite number of points in which they are not 1-manifolds. Otherwise, they are 1-manifolds in all extents. The combination presented, of PCA and V-D methods, allows the recovery of 1-manifold approximations for C-i(u) for self-intersecting quasi-planar and non self-intersecting curves. In the later case the PCA alone succeeds in finding 1-manifold PL approximations for them. The algorithm implemented finds applications in contour and shape reconstruction from noisy data, subject to sampling errors or blockage.Ítem Principal component analysis -PCA- and delone triangulations for PL approximation C-1-continuous 1-manifolds in R-N(ACTA PRESS, 2004-01-01) Ruiz, OE; Cadavid, CA; Garcia, MJ; Martinod, R; Ruiz, OE; Cadavid, CA; Garcia, MJ; Martinod, R; Universidad EAFIT. Departamento de Ingeniería Mecánica; Mecánica AplicadaA Method is presented which combines statistical (Principal Component Analysis) and deterministic (Voronoi-Delone) methods to find Piecewise Linear approximations of curves C-i(u) in R-3 sampled with statistical noise. If the curves are self-intersecting, there are a finite number of points in which they are not 1-manifolds. Otherwise, they are 1-manifolds in all extents. The combination presented, of PCA and V-D methods, allows the recovery of 1-manifold approximations for C-i(u) for self-intersecting quasi-planar and non self-intersecting curves. In the later case the PCA alone succeeds in finding 1-manifold PL approximations for them. The algorithm implemented finds applications in contour and shape reconstruction from noisy data, subject to sampling errors or blockage.Ítem Validation of two tools for the evaluation of changes in the attitudes of students in frequent situations in Palliative Care(Spanish Association of Anglo-American Studies, 2016-01-01) Krikorian, Alicia; Pablo Roman, Juan; Wenk, Roberto; Monti, CarolinaObjective: To determine the psychometric properties of two scales designed to examine attitudes regarding palliative care: Comfort Scale in Palliative Care (CSPC, Pereira et al.) and Tanatophobia Scale (TS, Merrill et al.) Method: Seventy-seven students who completed an online course on psychosocial aspects of palliative care offered by the Latin American Association of Palliative Care participated in the study. They also completed the scales before and after the course. Construct validity and reliability of the CSPC and the TS were assessed using a Principal Components Analysis, internal reliability coefficient and test-retest reliability. Further, comparative statistics between the pre-course and post-course results were obtained in order to determine changes in attitudes. Results: The Principal Components Analysis showed satisfactory fit to the data. 3 components were extracted: two for the CSPC and one for the TS, which explained 55.37[%] of the variance. Internal consistency coefficients were satisfactory in all cases and Cronbach’s Alphas were satisfactory for all the scales, particularly for the CSPC. Test-retest reliability in t1 and t2 was found to be non significant, indicating that measures were not related in time. Regarding pre-course/post-course comparisons, significant changes in comfort assisting patients (p = 0.004) and comfort assisting families (p = 0.001) following the course were identified, but changes in thanatophobia were non significant (p > 0.05). Conclusions: both scales are valid and reliable. Attitudes regarding the practice of palliative care and how they change, particularly regarding psychosocial issues, can be accurately measured using the examined scales. © 2016, Spanish Association of Anglo-American Studies. All rights reserved.