Previsão de diagramas de carga via Holt-Winters com três sazonalidades

Dissertação para a obtenção do grau de Mestre em Engenharia Electrotécnica Ramo de Energia

Método de desagregação temporal para a previsão da taxa de desemprego em Portugal

Dissertação de mestrado em Estatística

Sistema de apoio à decisão clínica: quality of life system

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação

Assessment of Autonomic Function by Phase Rectification of RRInterval Histogram Analysis in Chagas Disease

Background:In chronic Chagas disease (ChD), impairment of cardiac autonomic function bears prognostic implications. Phase‑rectification of RR-interval series isolates the sympathetic, acceleration phase (AC) and parasympathetic, deceleration phase (DC) influences on cardiac autonomic modulation.Objective:This study investigated heart rate variability (HRV) as a function of RR-interval to assess autonomic function in healthy and ChD subjects.Methods:Control (n = 20) and ChD (n = 20) groups were studied. All underwent 60-min head-up tilt table test under ECG recording. Histogram of RR-interval series was calculated, with 100 ms class, ranging from 600–1100 ms. In each class, mean RR-intervals (MNN) and root-mean-squared difference (RMSNN) of consecutive normal RR-intervals that suited a particular class were calculated. Average of all RMSNN values in each class was analyzed as function of MNN, in the whole series (RMSNNT), and in AC (RMSNNAC) and DC (RMSNNDC) phases. Slopes of linear regression lines were compared between groups using Student t-test. Correlation coefficients were tested before comparisons. RMSNN was log-transformed. (α < 0.05).Results:Correlation coefficient was significant in all regressions (p < 0.05). In the control group, RMSNNT, RMSNNAC, and RMSNNDCsignificantly increased linearly with MNN (p < 0.05). In ChD, only RMSNNAC showed significant increase as a function of MNN, whereas RMSNNT and RMSNNDC did not.Conclusion:HRV increases in proportion with the RR-interval in healthy subjects. This behavior is lost in ChD, particularly in the DC phase, indicating cardiac vagal incompetence.

Comparative Analysis of Signals Restoration by Different Kinds of Approximation

The comparative analysis of continuous signals restoration by different kinds of approximation is performed. The software product, allowing to define optimal method of different original signals restoration by Lagrange polynomial, Kotelnikov interpolation series, linear and cubic splines, Haar wavelet and Kotelnikov-Shannon wavelet based on criterion of minimum value of mean-square deviation is proposed. Practical recommendations on the selection of approximation function for different class of signals are obtained.

Forecasting US bond default ratings allowing for previous and initial state dependence in an ordered probit model

In this paper we investigate the ability of a number of different ordered probit models to predict ratings based on firm-specific data on business and financial risks. We investigate models based on momentum, drift and ageing and compare them against alternatives that take into account the initial rating of the firm and its previous actual rating. Using data on US bond issuing firms rated by Fitch over the years 2000 to 2007 we compare the performance of these models in predicting the rating in-sample and out-of-sample using root mean squared errors, Diebold-Mariano tests of forecast performance and contingency tables. We conclude that initial and previous states have a substantial influence on rating prediction.

Multivariate QST-FST comparisons: a neutrality test for the evolution of the g matrix in structured populations.

Neutrality tests in quantitative genetics provide a statistical framework for the detection of selection on polygenic traits in wild populations. However, the existing method based on comparisons of divergence at neutral markers and quantitative traits (Q(st)-F(st)) suffers from several limitations that hinder a clear interpretation of the results with typical empirical designs. In this article, we propose a multivariate extension of this neutrality test based on empirical estimates of the among-populations (D) and within-populations (G) covariance matrices by MANOVA. A simple pattern is expected under neutrality: D = 2F(st)/(1 - F(st))G, so that neutrality implies both proportionality of the two matrices and a specific value of the proportionality coefficient. This pattern is tested using Flury's framework for matrix comparison [common principal-component (CPC) analysis], a well-known tool in G matrix evolution studies. We show the importance of using a Bartlett adjustment of the test for the small sample sizes typically found in empirical studies. We propose a dual test: (i) that the proportionality coefficient is not different from its neutral expectation [2F(st)/(1 - F(st))] and (ii) that the MANOVA estimates of mean square matrices between and among populations are proportional. These two tests combined provide a more stringent test for neutrality than the classic Q(st)-F(st) comparison and avoid several statistical problems. Extensive simulations of realistic empirical designs suggest that these tests correctly detect the expected pattern under neutrality and have enough power to efficiently detect mild to strong selection (homogeneous, heterogeneous, or mixed) when it is occurring on a set of traits. This method also provides a rigorous and quantitative framework for disentangling the effects of different selection regimes and of drift on the evolution of the G matrix. We discuss practical requirements for the proper application of our test in empirical studies and potential extensions.

Error and complexity of random walk Monte Carlo radiosity

The author studies the error and complexity of the discrete random walk Monte Carlo technique for radiosity, using both the shooting and gathering methods. The author shows that the shooting method exhibits a lower complexity than the gathering one, and under some constraints, it has a linear complexity. This is an improvement over a previous result that pointed to an O(n log n) complexity. The author gives and compares three unbiased estimators for each method, and obtains closed forms and bounds for their variances. The author also bounds the expected value of the mean square error (MSE). Some of the results obtained are also shown

Effective diagnosis of Alzheimer's disease by means of large margin-based methodology.

BACKGROUND Functional brain images such as Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) have been widely used to guide the clinicians in the Alzheimer's Disease (AD) diagnosis. However, the subjectivity involved in their evaluation has favoured the development of Computer Aided Diagnosis (CAD) Systems. METHODS It is proposed a novel combination of feature extraction techniques to improve the diagnosis of AD. Firstly, Regions of Interest (ROIs) are selected by means of a t-test carried out on 3D Normalised Mean Square Error (NMSE) features restricted to be located within a predefined brain activation mask. In order to address the small sample-size problem, the dimension of the feature space was further reduced by: Large Margin Nearest Neighbours using a rectangular matrix (LMNN-RECT), Principal Component Analysis (PCA) or Partial Least Squares (PLS) (the two latter also analysed with a LMNN transformation). Regarding the classifiers, kernel Support Vector Machines (SVMs) and LMNN using Euclidean, Mahalanobis and Energy-based metrics were compared. RESULTS Several experiments were conducted in order to evaluate the proposed LMNN-based feature extraction algorithms and its benefits as: i) linear transformation of the PLS or PCA reduced data, ii) feature reduction technique, and iii) classifier (with Euclidean, Mahalanobis or Energy-based methodology). The system was evaluated by means of k-fold cross-validation yielding accuracy, sensitivity and specificity values of 92.78%, 91.07% and 95.12% (for SPECT) and 90.67%, 88% and 93.33% (for PET), respectively, when a NMSE-PLS-LMNN feature extraction method was used in combination with a SVM classifier, thus outperforming recently reported baseline methods. CONCLUSIONS All the proposed methods turned out to be a valid solution for the presented problem. One of the advances is the robustness of the LMNN algorithm that not only provides higher separation rate between the classes but it also makes (in combination with NMSE and PLS) this rate variation more stable. In addition, their generalization ability is another advance since several experiments were performed on two image modalities (SPECT and PET).

Adaptively weighted maximum likelihood estimation of discrete distributions

SummaryDiscrete data arise in various research fields, typically when the observations are count data.I propose a robust and efficient parametric procedure for estimation of discrete distributions. The estimation is done in two phases. First, a very robust, but possibly inefficient, estimate of the model parameters is computed and used to indentify outliers. Then the outliers are either removed from the sample or given low weights, and a weighted maximum likelihood estimate (WML) is computed.The weights are determined via an adaptive process such that if the data follow the model, then asymptotically no observation is downweighted.I prove that the final estimator inherits the breakdown point of the initial one, and that its influence function at the model is the same as the influence function of the maximum likelihood estimator, which strongly suggests that it is asymptotically fully efficient.The initial estimator is a minimum disparity estimator (MDE). MDEs can be shown to have full asymptotic efficiency, and some MDEs have very high breakdown points and very low bias under contamination. Several initial estimators are considered, and the performances of the WMLs based on each of them are studied.It results that in a great variety of situations the WML substantially improves the initial estimator, both in terms of finite sample mean square error and in terms of bias under contamination. Besides, the performances of the WML are rather stable under a change of the MDE even if the MDEs have very different behaviors.Two examples of application of the WML to real data are considered. In both of them, the necessity for a robust estimator is clear: the maximum likelihood estimator is badly corrupted by the presence of a few outliers.This procedure is particularly natural in the discrete distribution setting, but could be extended to the continuous case, for which a possible procedure is sketched.RésuméLes données discrètes sont présentes dans différents domaines de recherche, en particulier lorsque les observations sont des comptages.Je propose une méthode paramétrique robuste et efficace pour l'estimation de distributions discrètes. L'estimation est faite en deux phases. Tout d'abord, un estimateur très robuste des paramètres du modèle est calculé, et utilisé pour la détection des données aberrantes (outliers). Cet estimateur n'est pas nécessairement efficace. Ensuite, soit les outliers sont retirés de l'échantillon, soit des faibles poids leur sont attribués, et un estimateur du maximum de vraisemblance pondéré (WML) est calculé.Les poids sont déterminés via un processus adaptif, tel qu'asymptotiquement, si les données suivent le modèle, aucune observation n'est dépondérée.Je prouve que le point de rupture de l'estimateur final est au moins aussi élevé que celui de l'estimateur initial, et que sa fonction d'influence au modèle est la même que celle du maximum de vraisemblance, ce qui suggère que cet estimateur est pleinement efficace asymptotiquement.L'estimateur initial est un estimateur de disparité minimale (MDE). Les MDE sont asymptotiquement pleinement efficaces, et certains d'entre eux ont un point de rupture très élevé et un très faible biais sous contamination. J'étudie les performances du WML basé sur différents MDEs.Le résultat est que dans une grande variété de situations le WML améliore largement les performances de l'estimateur initial, autant en terme du carré moyen de l'erreur que du biais sous contamination. De plus, les performances du WML restent assez stables lorsqu'on change l'estimateur initial, même si les différents MDEs ont des comportements très différents.Je considère deux exemples d'application du WML à des données réelles, où la nécessité d'un estimateur robuste est manifeste : l'estimateur du maximum de vraisemblance est fortement corrompu par la présence de quelques outliers.La méthode proposée est particulièrement naturelle dans le cadre des distributions discrètes, mais pourrait être étendue au cas continu.

Optimum power allocation for parallel Gaussian channels with arbitrary input distributions

The mutual information of independent parallel Gaussian-noise channels is maximized, under an average power constraint, by independent Gaussian inputs whose power is allocated according to the waterfilling policy. In practice, discrete signalling constellations with limited peak-to-average ratios (m-PSK, m-QAM, etc) are used in lieu of the ideal Gaussian signals. This paper gives the power allocation policy that maximizes the mutual information over parallel channels with arbitrary input distributions. Such policy admits a graphical interpretation, referred to as mercury/waterfilling, which generalizes the waterfilling solution and allows retaining some of its intuition. The relationship between mutual information of Gaussian channels and nonlinear minimum mean-square error proves key to solving the power allocation problem.

Large-system analysis of a CDMA dynamic channel under a Markovian input process

We study the minimum mean square error (MMSE) and the multiuser efficiency η of large dynamic multiple access communication systems in which optimal multiuser detection is performed at the receiver as the number and the identities of active users is allowed to change at each transmission time. The system dynamics are ruled by a Markov model describing the evolution of the channel occupancy and a large-system analysis is performed when the number of observations grow large. Starting on the equivalent scalar channel and the fixed-point equation tying multiuser efficiency and MMSE, we extend it to the case of a dynamic channel, and derive lower and upper bounds for the MMSE (and, thus, for η as well) holding true in the limit of large signal–to–noise ratios and increasingly large observation time T.

Surface deterioration of dental materials after simulated toothbrushing in relation to brushing time and load.

OBJECTIVES: (1) To evaluate the changes in surface roughness and gloss after simulated toothbrushing of 9 composite materials and 2 ceramic materials in relation to brushing time and load in vitro; (2) to assess the relationship between surface gloss and surface roughness. METHODS: Eight flat specimens of composite materials (microfilled: Adoro, Filtek Supreme, Heliomolar; microhybrid: Four Seasons, Tetric EvoCeram; hybrid: Compoglass F, Targis, Tetric Ceram; macrohybrid: Grandio), two ceramic materials (IPS d.SIGN and IPS Empress polished) were fabricated according to the manufacturer's instructions and optimally polished with up to 4000 grit SiC. The specimens were subjected to a toothbrushing (TB) simulation device (Willytec) with rotating movements, toothpaste slurry and at three different loads (100g/250g/350g). At hourly intervals from 1h to 10h TB, mean surface roughness Ra was measured with an optical sensor and the surface gloss (Gl) with a glossmeter. Statistical analysis was performed for log-transformed Ra data applying two-way ANOVA to evaluate the interaction between load and material and load and brushing time. RESULTS: There was a significant interaction between material and load as well as between load and brushing time (p<0.0001). The microhybrid and hybrid materials demonstrated more surface deterioration with higher loads, whereas with the microfilled resins Heliomolar and Adoro it was vice versa. For ceramic materials, no or little deterioration was observed over time and independent of the load. The ceramic materials and 3 of the composite materials (roughness) showed no further deterioration after 5h of toothbrushing. Mean surface gloss was the parameter which discriminated best between the materials, followed by mean surface roughness Ra. There was a strong correlation between surface gloss and surface roughness for all the materials except the ceramics. The evaluation of the deterioration curves of individual specimens revealed a more or less synchronous course suspecting hinting specific external conditions and not showing the true variability in relation to the tested material. SIGNIFICANCE: The surface roughness and gloss of dental materials changes with brushing time and load and thus results in different material rankings. Apart from Grandio, the hybrid composite resins were more prone to surface changes than microfilled composites. The deterioration potential of a composite material can be quickly assessed by measuring surface gloss. For this purpose, a brushing time of 10h (=72,000 strokes) is needed. In further comparative studies, specimens of different materials should be tested in one series to estimate the true variability.

Using composite estimators to improve both domain and total area estimation

In this article we propose using small area estimators to improve the estimatesof both the small and large area parameters. When the objective is to estimateparameters at both levels accurately, optimality is achieved by a mixed sampledesign of fixed and proportional allocations. In the mixed sample design, oncea sample size has been determined, one fraction of it is distributedproportionally among the different small areas while the rest is evenlydistributed among them. We use Monte Carlo simulations to assess theperformance of the direct estimator and two composite covariant-freesmall area estimators, for different sample sizes and different sampledistributions. Performance is measured in terms of Mean Squared Errors(MSE) of both small and large area parameters. It is found that the adoptionof small area composite estimators open the possibility of 1) reducingsample size when precision is given, or 2) improving precision for a givensample size.

Forecasting with missing data: Application to a real case

This paper presents a comparative analysis of linear and mixed modelsfor short term forecasting of a real data series with a high percentage of missing data. Data are the series of significant wave heights registered at regular periods of three hours by a buoy placed in the Bay of Biscay.The series is interpolated with a linear predictor which minimizes theforecast mean square error. The linear models are seasonal ARIMA models and themixed models have a linear component and a non linear seasonal component.The non linear component is estimated by a non parametric regression of dataversus time. Short term forecasts, no more than two days ahead, are of interestbecause they can be used by the port authorities to notice the fleet.Several models are fitted and compared by their forecasting behavior.