Análisis diacrónico de la producción española de tesis doctorales en educación matemática mediante la metodología ARIMA (Auto-Regressive Integrated Moving Average) en datos de diseños longitudinales.

Se analiza la manera en que se realizan las tesis doctorales en educación matemática en España. Se utiliza la metodología ARIMA (Auto-Regressive Integrated Moving Average) para realizar el análisis de manera diacrónica sobre datos longitudinales. Se hace incapié en la importancia de la metodología usada y sus ventajas frente a las metodologías tradicionalmente usadas en análisis diacrónicos. Se exponen las cuatro fases de la metodología ARIMA, correspondientes a la identificación del proceso, la estimación de cambio en el proceso, la validación del mismo y la predicción de sus consecuencias.

Application of auto regressive models of wavelet sub-bands for classifying Terahertz pulse measurements

This paper presents an approach for automatic classification of pulsed Terahertz (THz), or T-ray, signals highlighting their potential in biomedical, pharmaceutical and security applications. T-ray classification systems supply a wealth of information about test samples and make possible the discrimination of heterogeneous layers within an object. In this paper, a novel technique involving the use of Auto Regressive (AR) and Auto Regressive Moving Average (ARMA) models on the wavelet transforms of measured T-ray pulse data is presented. Two example applications are examined - the classi. cation of normal human bone (NHB) osteoblasts against human osteosarcoma (HOS) cells and the identification of six different powder samples. A variety of model types and orders are used to generate descriptive features for subsequent classification. Wavelet-based de-noising with soft threshold shrinkage is applied to the measured T-ray signals prior to modeling. For classi. cation, a simple Mahalanobis distance classi. er is used. After feature extraction, classi. cation accuracy for cancerous and normal cell types is 93%, whereas for powders, it is 98%.

Um modelo estocástico combinado de previsão sazonal para a precipitação no Brasil

Este artigo discute um modelo de previsão combinada para a realização de prognósticos climáticos na escala sazonal. Nele, previsões pontuais de modelos estocásticos são agregadas para obter as melhores projeções no tempo. Utilizam-se modelos estocásticos autoregressivos integrados a médias móveis, de suavização exponencial e previsões por análise de correlações canônicas. O controle de qualidade das previsões é feito através da análise dos resíduos e da avaliação do percentual de redução da variância não-explicada da modelagem combinada em relação às previsões dos modelos individuais. Exemplos da aplicação desses conceitos em modelos desenvolvidos no Instituto Nacional de Meteorologia (INMET) mostram bons resultados e ilustram que as previsões do modelo combinado, superam na maior parte dos casos a de cada modelo componente, quando comparadas aos dados observados.

Análise espacial dos determinantes socioeconômicos dos homicídios no Estado de Pernambuco

OBJETIVO: Investigar a associação entre variáveis socioeconômicas e taxas de homicídio, considerando a localização espacial dos indicadores. MÉTODOS: Utilizou-se o método de estudo ecológico. A variável dependente foi taxa de homicídio da população masculina de 15 a 49 anos, residente nos municípios do Estado de Pernambuco, em 1995 a 1998. As variáveis independentes referem-se a: índice de condições de vida, renda familiar per capita, desigualdade de Theil, índice de Gini, renda média do chefe de família, índice de pobreza, taxa de analfabetismo, densidade demográfica.Utilizou-se teste de correlação espacial determinado pelo Índice de Moran, regressão múltipla, Conditional Auto Regressive (CAR) e a função Loess, como modelo de detecção de tendência especial. RESULTADOS: Os indicadores taxa de analfabetismo e índice de pobreza explicaram 24,6% da variabilidade total das taxas de homicídio, cuja associação foi inversa. O índice de Moran revelou autocorrelação espacial entre os municípios. O modelo de regressão espacial que melhor se adequou ao estudo foi o CAR, que confirmou a associação entre índice de pobreza, analfabetismo e homicídio. CONCLUSÕES: A relação inversa observada entre os indicadores socioeconômicos e homicídios pode expressar determinado processo que propicia melhoria das condições de vida, e está atrelado predominantemente a condições geradoras de violência, como a do tráfico de drogas.

Seasonal variation in hospitalizations due to heart failure in Niterói city, Southeastern Brazil

The objective of the study was to describe seasonality of hospitalizations for heart failure in tropical climate as it has been described in cold climates. Seasonal Auto-regressive Integrated Moving-Average model was applied to time-series data of heart failure hospitalizations between 1996 and 2004 in Niteroi (Southeastern Brazil), collected from the Brazilian National Health Service Database. The standard seasonal variation was obtained by means of moving-average filtering and averaging data. The lowest and the highest annual hospital admissions were 507 (1997) and 849 (2002), respectively; the lowest and the highest monthly rates were 419 (December) and 681 (October), respectively. Peak admission rates were seen during the fall and winter. Although weak, the seasonality observed indicates that slight variations result in increased hospitalizations for heart failure.

A new least-squares approach to differintegration modeling

Signal Processing, Vol. 86, nº 10

Inserção de todos os consumidores no mercado liberalizado de energia elétrica: oportunidades e desafios

O processo de liberalização do setor elétrico em Portugal Continental seguiu uma metodologia idêntica à da maior parte dos países europeus, tendo a abertura de mercado sido efetuada de forma progressiva. Assim, no âmbito do acompanhamento do setor elétrico nacional, reveste-se de particular interesse caracterizar a evolução mais recente do mercado liberalizado, nomeadamente em relação ao preço da energia elétrica. A previsão do preço da energia elétrica é uma questão muito importante para todos os participantes do mercado de energia elétrica. Como se trata de um assunto de grande importância, a previsão do preço da energia elétrica tem sido alvo de diversos estudos e diversas metodologias têm sido propostas. Esta questão é abordada na presente dissertação recorrendo a técnicas de previsão, nomeadamente a métodos baseados no histórico da variável em estudo. As previsões são, segundo alguns especialistas, um dos inputs essenciais que os gestores desenvolvem para ajudar no processo de decisão. Virtualmente cada decisão relevante ao nível das operações depende de uma previsão. Para a realização do modelo de previsão de preço da energia elétrica foram utilizados os modelos Autorregressivos Integrados de Médias Móveis, Autoregressive / Integrated / Moving Average (ARIMA), que geram previsões através da informação contida na própria série temporal. Como se pretende avaliar a estrutura do preço da energia elétrica do mercado de energia, é importante identificar, deste conjunto de variáveis, quais as que estão mais relacionados com o preço. Neste sentido, é realizada em paralelo uma análise exploratória, através da correlação entre o preço da energia elétrica e outras variáveis de estudo, utilizando para esse efeito o coeficiente de correlação de Pearson. O coeficiente de correlação de Pearson é uma medida do grau e da direção de relação linear entre duas variáveis quantitativas. O modelo desenvolvido foi aplicado tendo por base o histórico de preço da eletricidade desde o inicio do mercado liberalizado e de modo a obter as previsões diária, mensal e anual do preço da eletricidade. A metodologia desenvolvida demonstrou ser eficiente na obtenção das soluções e ser suficientemente rápida para prever o valor do preço da energia elétrica em poucos segundos, servindo de apoio à decisão em ambiente de mercado.

Application of wavelets to the study of political history

\The idea that social processes develop in a cyclical manner is somewhat like a `Lorelei'. Researchers are lured to it because of its theoretical promise, only to become entangled in (if not wrecked by) messy problems of empirical inference. The reasoning leading to hypotheses of some kind of cycle is often elegant enough, yet the data from repeated observations rarely display the supposed cyclical pattern. (...) In addition, various `schools' seem to exist which frequently arrive at di erent conclusions on the basis of the same data." (van der Eijk and Weber 1987:271). Much of the empirical controversies around these issues arise because of three distinct problems: the coexistence of cycles of di erent periodicities, the possibility of transient cycles and the existence of cycles without xed periodicity. In some cases, there are no reasons to expect any of these phenomena to be relevant. Seasonality caused by Christmas is one such example (Wen 2002). In such cases, researchers mostly rely on spectral analysis and Auto-Regressive Moving-Average (ARMA) models to estimate the periodicity of cycles.1 However, and this is particularly true in social sciences, sometimes there are good theoretical reasons to expect irregular cycles. In such cases, \the identi cation of periodic movement in something like the vote is a daunting task all by itself. When a pendulum swings with an irregular beat (frequency), and the extent of the swing (amplitude) is not constant, mathematical functions like sine-waves are of no use."(Lebo and Norpoth 2007:73) In the past, this di culty has led to two di erent approaches. On the one hand, some researchers dismissed these methods altogether, relying on informal alternatives that do not meet rigorous standards of statistical inference. Goldstein (1985 and 1988), studying the severity of Great power wars is one such example. On the other hand, there are authors who transfer the assumptions of spectral analysis (and ARMA models) into fundamental assumptions about the nature of social phenomena. This type of argument was produced by Beck (1991) who, in a reply to Goldstein (1988), claimed that only \ xed period models are meaningful models of cyclic phenomena".We argue that wavelet analysis|a mathematical framework developed in the mid-1980s (Grossman and Morlet 1984; Goupillaud et al. 1984) | is a very viable alternative to study cycles in political time-series. It has the advantage of staying close to the frequency domain approach of spectral analysis while addressing its main limitations. Its principal contribution comes from estimating the spectral characteristics of a time-series as a function of time, thus revealing how its di erent periodic components may change over time. The rest of article proceeds as follows. In the section \Time-frequency Analysis", we study in some detail the continuous wavelet transform and compare its time-frequency properties with the more standard tool for that purpose, the windowed Fourier transform. In the section \The British Political Pendulum", we apply wavelet analysis to essentially the same data analyzed by Lebo and Norpoth (2007) and Merrill, Grofman and Brunell (2011) and try to provide a more nuanced answer to the same question discussed by these authors: do British electoral politics exhibit cycles? Finally, in the last section, we present a concise list of future directions.

The Difference, System and ‘Double-D’ GMM Panel Estimators in the Presence of Structural Breaks

The effects of structural breaks in dynamic panels are more complicated than in time series models as the bias can be either negative or positive. This paper focuses on the effects of mean shifts in otherwise stationary processes within an instrumental variable panel estimation framework. We show the sources of the bias and a Monte Carlo analysis calibrated on United States bank lending data demonstrates the size of the bias for a range of auto-regressive parameters. We also propose additional moment conditions that can be used to reduce the biases caused by shifts in the mean of the data.

A temporal hierarchy for conspecific vocalization discrimination in humans.

Introduction: Discrimination of species-specific vocalizations is fundamental for survival and social interactions. Its unique behavioral relevance has encouraged the identification of circumscribed brain regions exhibiting selective responses (Belin et al., 2004), while the role of network dynamics has received less attention. Those studies that have examined the brain dynamics of vocalization discrimination leave unresolved the timing and the inter-relationship between general categorization, attention, and speech-related processes (Levy et al., 2001, 2003; Charest et al., 2009). Given these discrepancies and the presence of several confounding factors, electrical neuroimaging analyses were applied to auditory evoked-potential (AEPs) to acoustically and psychophysically controlled non-verbal human and animal vocalizations. This revealed which region(s) exhibit voice-sensitive responses and in which sequence. Methods: Subjects (N=10) performed a living vs. man-made 'oddball' auditory discrimination task, such that on a given block of trials 'target' stimuli occurred 10% of the time. Stimuli were complex, meaningful sounds of 500ms duration. There were 120 different sound files in total, 60 of which represented sounds of living objects and 60 man-made objects. The stimuli that were the focus of the present investigation were restricted to those of living objects within blocks where no response was required. These stimuli were further sorted between human non-verbal vocalizations and animal vocalizations. They were also controlled in terms of their spectrograms and formant distributions. Continuous 64-channel EEG was acquired through Neuroscan Synamps referenced to the nose, band-pass filtered 0.05-200Hz, and digitized at 1000Hz. Peri-stimulus epochs of continuous EEG (-100ms to 900ms) were visually inspected for artifacts, 40Hz low-passed filtered and baseline corrected using the pre-stimulus period . Averages were computed from each subject separately. AEPs in response to animal and human vocalizations were analyzed with respect to differences of Global Field Power (GFP) and with respect to changes of the voltage configurations at the scalp (reviewed in Murray et al., 2008). The former provides a measure of the strength of the electric field irrespective of topographic differences; the latter identifies changes in spatial configurations of the underlying sources independently of the response strength. In addition, we utilized the local auto-regressive average distributed linear inverse solution (LAURA; Grave de Peralta Menendez et al., 2001) to visualize and statistically contrast the likely underlying sources of effects identified in the preceding analysis steps. Results: We found differential activity in response to human vocalizations over three periods in the post-stimulus interval, and this response was always stronger than that to animal vocalizations. The first differential response (169-219ms) was a consequence of a modulation in strength of a common brain network localized into the right superior temporal sulcus (STS; Brodmann's Area (BA) 22) and extending into the superior temporal gyrus (STG; BA 41). A second difference (291-357ms) also followed from strength modulations of a common network with statistical differences localized to the left inferior precentral and prefrontal gyrus (BA 6/45). These two first strength modulations correlated (Spearman's rho(8)=0.770; p=0.009) indicative of functional coupling between temporally segregated stages of vocalization discrimination. A third difference (389-667ms) followed from strength and topographic modulations and was localized to the left superior frontal gyrus (BA10) although this third difference did not reach our spatial criterion of 12 continuous voxels. Conclusions: We show that voice discrimination unfolds over multiple temporal stages, involving a wide network of brain regions. The initial stages of vocalization discrimination are based on modulations in response strength within a common brain network with no evidence for a voice-selective module. The latency of this effect parallels that of face discrimination (Bentin et al., 2007), supporting the possibility that voice and face processes can mutually inform one another. Putative underlying sources (localized in the right STS; BA 22) are consistent with prior hemodynamic imaging evidence in humans (Belin et al., 2004). Our effect over the 291-357ms post-stimulus period overlaps the 'voice-specific-response' reported by Levy et al. (Levy et al., 2001) and the estimated underlying sources (left BA6/45) were in agreement with previous findings in humans (Fecteau et al., 2005). These results challenge the idea that circumscribed and selective areas subserve con-specific vocalization processing.

State-space approach to evaluate the relation between soil physical and chemical properties

The state-space approach is used to evaluate the relation between soil physical and chemical properties in an area cultivated with sugarcane. The experiment was carried out on a Rhodic Kandiudalf in Piracicaba, State of São Paulo, Brazil. Sugarcane was planted on an area of 0.21 ha i.e., in 15 rows 100 m long, spaced 1.4 m. Soil water content, soil organic matter, clay content and aggregate stability were sampled along a transect of 84 points, meter by meter. The state-space approach is used to evaluate how the soil water content is affected by itself and by soil organic matter, clay content, and aggregate stability of neighboring locations, in different combinations, aiming to contribute to a better understanding of the relation among these variables in the soil. Results show that soil water contents were successfully estimated by this approach. Best performances were found when the estimate of soil water content at locations i was related to soil water content, clay content and aggregate stability at locations i-1. Results also indicate that this state-space model using all series describes the soil water content better than any equivalent multiple regression equation.

Grabbing your ear: rapid auditory-somatosensory multisensory interactions in low-level sensory cortices are not constrained by stimulus alignment.

Multisensory interactions are observed in species from single-cell organisms to humans. Important early work was primarily carried out in the cat superior colliculus and a set of critical parameters for their occurrence were defined. Primary among these were temporal synchrony and spatial alignment of bisensory inputs. Here, we assessed whether spatial alignment was also a critical parameter for the temporally earliest multisensory interactions that are observed in lower-level sensory cortices of the human. While multisensory interactions in humans have been shown behaviorally for spatially disparate stimuli (e.g. the ventriloquist effect), it is not clear if such effects are due to early sensory level integration or later perceptual level processing. In the present study, we used psychophysical and electrophysiological indices to show that auditory-somatosensory interactions in humans occur via the same early sensory mechanism both when stimuli are in and out of spatial register. Subjects more rapidly detected multisensory than unisensory events. At just 50 ms post-stimulus, neural responses to the multisensory 'whole' were greater than the summed responses from the constituent unisensory 'parts'. For all spatial configurations, this effect followed from a modulation of the strength of brain responses, rather than the activation of regions specifically responsive to multisensory pairs. Using the local auto-regressive average source estimation, we localized the initial auditory-somatosensory interactions to auditory association areas contralateral to the side of somatosensory stimulation. Thus, multisensory interactions can occur across wide peripersonal spatial separations remarkably early in sensory processing and in cortical regions traditionally considered unisensory.

Regression analysis of spatial data.

Many of the most interesting questions ecologists ask lead to analyses of spatial data. Yet, perhaps confused by the large number of statistical models and fitting methods available, many ecologists seem to believe this is best left to specialists. Here, we describe the issues that need consideration when analysing spatial data and illustrate these using simulation studies. Our comparative analysis involves using methods including generalized least squares, spatial filters, wavelet revised models, conditional autoregressive models and generalized additive mixed models to estimate regression coefficients from synthetic but realistic data sets, including some which violate standard regression assumptions. We assess the performance of each method using two measures and using statistical error rates for model selection. Methods that performed well included generalized least squares family of models and a Bayesian implementation of the conditional auto-regressive model. Ordinary least squares also performed adequately in the absence of model selection, but had poorly controlled Type I error rates and so did not show the improvements in performance under model selection when using the above methods. Removing large-scale spatial trends in the response led to poor performance. These are empirical results; hence extrapolation of these findings to other situations should be performed cautiously. Nevertheless, our simulation-based approach provides much stronger evidence for comparative analysis than assessments based on single or small numbers of data sets, and should be considered a necessary foundation for statements of this type in future.

Source separation techniques applied to linear prediction

The prediction filters are well known models for signal estimation, in communications, control and many others areas. The classical method for deriving linear prediction coding (LPC) filters is often based on the minimization of a mean square error (MSE). Consequently, second order statistics are only required, but the estimation is only optimal if the residue is independent and identically distributed (iid) Gaussian. In this paper, we derive the ML estimate of the prediction filter. Relationships with robust estimation of auto-regressive (AR) processes, with blind deconvolution and with source separation based on mutual information minimization are then detailed. The algorithm, based on the minimization of a high-order statistics criterion, uses on-line estimation of the residue statistics. Experimental results emphasize on the interest of this approach.

Parameter identification in time series models

Time series analysis can be categorized into three different approaches: classical, Box-Jenkins, and State space. Classical approach makes a basement for the analysis and Box-Jenkins approach is an improvement of the classical approach and deals with stationary time series. State space approach allows time variant factors and covers up a broader area of time series analysis. This thesis focuses on parameter identifiablity of different parameter estimation methods such as LSQ, Yule-Walker, MLE which are used in the above time series analysis approaches. Also the Kalman filter method and smoothing techniques are integrated with the state space approach and MLE method to estimate parameters allowing them to change over time. Parameter estimation is carried out by repeating estimation and integrating with MCMC and inspect how well different estimation methods can identify the optimal model parameters. Identification is performed in probabilistic and general senses and compare the results in order to study and represent identifiability more informative way.