Repository logo
 
Loading...
Project Logo
Research Project

Untitled

Authors

Publications

Time series analysis of data for sea surface temperature and upwelling components from the southwest coast of Portugal
Publication . Costa Goela, Priscila; Cordeiro, Clara; Danchenko, Sergei; Icely, John; Cristina, Sónia; Newton, Alice
This study relates sea surface temperature (SST) to the upwelling conditions off the southwest coast of Portugal using statistical analyses of publically available data. Optimum Interpolation (OI) of daily SST data were extracted from the United States (US) National Oceanic and Atmospheric Administration (NOAA) and data for wind speed and direction were from the US National Climatic Data Center. Time series were extracted at a daily frequency for a time horizon of 26 years. Upwelling indices were estimated using westerly (Q(x)) and southerly (Q(y)) Ekman transport components.In the first part of the study, time series were inspected for trend and seasonality over the whole period. The seasonally adjusted time series revealed an increasing slope for SST (0.15 degrees C per decade) and decreasing slopes for Q(x) (84.01 m(3) s(-1) km(-1) per decade) and Q(y) (-25.20 m(3) s(-1) km(-1) per decade), over the time horizon. Structural breaks analysis applied to the time series showed that a statistically significant incremental increase in SST was more pronounced during the last decade.Cross -correlation between upwelling indices and SST revealed a time delay of 5 and 2 days between Q(x) and SST, and between Qv and SST, respectively. A spectral analysis combined with the previous analysis enabled the identification of four oceanographic seasons. Those seasons were later recognised over a restricted time period of 4 years, between 2008 and 2012, when there was an extensive sampling programme for the validation of ocean colour remote sensing imagery. The seasons were defined as: summer, with intense and regular events of upwelling; autumn, indicating relaxation of upwelling conditions; and spring and winter, showing high inter annual variability in terms of number and intensity of upwelling events. (C) 2016 The Authors. Published by Elsevier B.V.
Item response theory. A first approach
Publication . Nunes, Sandra; Oliveira, Teresa; Oliveira, Amílcar
The Item Response Theory (IRT) has become one of the most popular scoring frameworks for measurement data, frequently used in computerized adaptive testing, cognitively diagnostic assessment and test equating. According to Andrade et al. (2000), IRT can be defined as a set of mathematical models (Item Response Models – IRM) constructed to represent the probability of an individual giving the right answer to an item of a particular test. The number of Item Responsible Models available to measurement analysis has increased considerably in the last fifteen years due to increasing computer power and due to a demand for accuracy and more meaningful inferences grounded in complex data. The developments in modeling with Item Response Theory were related with developments in estimation theory, most remarkably Bayesian estimation with Markov chain Monte Carlo algorithms (Patz & Junker, 1999). The popularity of Item Response Theory has also implied numerous overviews in books and journals, and many connections between IRT and other statistical estimation procedures, such as factor analysis and structural equation modeling, have been made repeatedly (Van der Lindem & Hambleton, 1997). As stated before the Item Response Theory covers a variety of measurement models, ranging from basic one-dimensional models for dichotomously and polytomously scored items and their multidimensional analogues to models that incorporate information about cognitive sub-processes which influence the overall item response process. The aim of this work is to introduce the main concepts associated with one-dimensional models of Item Response Theory, to specify the logistic models with one, two and three parameters, to discuss some properties of these models and to present the main estimation procedures.
MERIS phytoplankton time series products from the SW Iberian Peninsula (Sagres) using seasonal-trend decomposition based on loess
Publication . Cristina, Sónia; Cordeiro, Clara; Lavender, Samantha; Goela, Priscila; Icely, John; Newton, Alice
The European Space Agency has acquired 10 years of data on the temporal and spatial distribution of phytoplankton biomass from the MEdium Resolution Imaging Spectrometer (MERIS) sensor for ocean color. The phytoplankton biomass was estimated with the MERIS product Algal Pigment Index 1 (API 1). Seasonal-Trend decomposition of time series based on Loess (STL) identified the temporal variability of the dynamical features in the MERIS products for water leaving reflectance ((w)()) and API 1. The advantages of STL is that it can identify seasonal components changing over time, it is responsive to nonlinear trends, and it is robust in the presence of outliers. One of the novelties in this study is the development and the implementation of an automatic procedure, stl.fit(), that searches the best data modeling by varying the values of the smoothing parameters, and by selecting the model with the lowest error measure. This procedure was applied to 10 years of monthly time series from Sagres in the Southwestern Iberian Peninsula at three Stations, 2, 10 and 18 km from the shore. Decomposing the MERIS products into seasonal, trend and irregular components with stl.fit(), the (w)() indicated dominance of the seasonal and irregular components while API 1 was mainly dominated by the seasonal component, with an increasing effect from inshore to offshore. A comparison of the seasonal components between the (w)() and the API 1 product, showed that the variations decrease along this time period due to the changes in phytoplankton functional types. Furthermore, inter-annual seasonal variation for API 1 showed the influence of upwelling events and in which month of the year these occur at each of the three Sagres stations. The stl.fit() is a good tool for any remote sensing study of time series, particularly those addressing inter-annual variations. This procedure will be made available in R software.
The presence of distortions in the extended skew: normal distribution
Publication . Seijas-Macias, António; Oliveira, Amílcar; Oliveira, Teresa
In the last years, a very interesting topic has arisen and became the research focus not only for many mathematicians and statisticians, as well as for all those interested in modeling issues: The Skew normal distributions’ family that represents a generalization of normal distribution. The first generalization was developed by Azzalini in 1985, which produces the skew-normal distribution, and introduces the existence of skewness into the normal distribution. Later on, the extended skew-normal distribution is defined as a generalization of skew-normal distribution. These distributions are potentially useful for the data that presenting high values of skewness and kurtosis. Applications of this type of distributions are very common in model of economic data, especially when asymmetric models are underlying the data. Definition of this type of distribution is based in four parameters: location, scale, shape and truncation. In this paper, we analyze the evolution of skewness and kurtosis of extended skew-normal distribution as a function of two parameters (shape and truncation). We focus in the value of kurtosis and skewness and the existence of arange of values where tiny modification of the parameters produces large oscillations in the values. The analysis shows that skewness and kurtosis present an instability development for greater values of truncation. Moreover, some values of kurtosis could be erroneous. Packages implemented in software R confirm the existence of a range where value of kurtosis presents a random evolution.

Organizational Units

Description

Keywords

Contributors

Funders

Funding agency

Fundação para a Ciência e a Tecnologia

Funding programme

5876

Funding Award Number

UID/MAT/00006/2013

ID