Comparison for three Classification Techniques

In this paper we show the results of a comparison simulation study for three classification techniques: Multinomial Logistic Regression (MLR), No Metric Discriminant Analysis (NDA) and Linear Discriminant Analysis (LDA). The measure used to compare the performance of the three techniques was the Error Classification Rate (ECR). We found that MLR and LDA techniques have similar performance and that they are better than DNA when the population multivariate distribution is Normal or Logit-Normal. For the case of log-normal and Sinh(-1)-normal multivariate distributions we found that MLR had the better performance.

Endogenous normative indices for the industry concentration

This paper presents new indices for measuring the industry concentration. The indices proposed (C n ) are of a normative type because they embody (endogenous) weights matching the market shares of the individual firms to their Marshallian welfare shares. These indices belong to an enlarged class of the Performance Gradient Indexes introduced by Dansby&Willig(I979). The definition of Cn for the consumers allows a new interpretation for the Hirschman-Herfindahl index (H), which can be viewed as a normative index according to particular values of the demand parameters. For homogeneous product industries, Cn equates H for every market distribution if (and only if) the market demand is linear. Whenever the inverse demand curve is convex (concave), H underestimates( overestimates) the industry concentration measured by the normative indexo For these industries, H overestimates (underestimates) the concentration changes caused by market transfers among small firms if the inverse demand curve is convex(concave) and underestimates( overestimates) it when such tranfers benefit a large firm, according to the convexity (or the concavity) of the demand curve. For heterogeneous product industries, an explicit normative index is obtained with a market demand derived from a quasi-linear utilility function. Under symmetric preferences among the goods, the index Cn is always greater than or equal the H-index. Under asymmetric assumptions, discrepancies between the firms' market distribution and the differentiationj substitution distributions among the goods, increase the concentration but make room for some horizontal mergers do reduce it. In particular, a mean preserving spread of the differentiation(substitution) increases(decreases) the concentration only if the smaller firms' goods become more(less) differentiated(substitute) w.r.t. the other goods. One important consequence of these results is that the consumers are benefitted when the smaller firms are producing weak substitute goods, and the larger firms produce strong substitute goods or face demand curves weakly sensitive to their own prices.

Multivariate analysis for the classification and differentiation of Madeira wines according to main grape varieties

In order to differentiate and characterize Madeira wines according to main grape varieties, the volatile composition (higher alcohols, fatty acids, ethyl esters and carbonyl compounds) was determined for 36 monovarietal Madeira wine samples elaborated from Boal, Malvazia, Sercial and Verdelho white grape varieties. The study was carried out by headspace solid-phase microextraction technique (HS-SPME), in dynamic mode, coupled with gas chromatography–mass spectrometry (GC–MS). Corrected peak area data for 42 analytes from the above mentioned chemical groups was used for statistical purposes. Principal component analysis (PCA) was applied in order to determine the main sources of variability present in the data sets and to establish the relation between samples (objects) and volatile compounds (variables). The data obtained by GC–MS shows that the most important contributions to the differentiation of Boal wines are benzyl alcohol and (E)-hex-3-en-1-ol. Ethyl octadecanoate, (Z)-hex-3-en-1-ol and benzoic acid are the major contributions in Malvazia wines and 2-methylpropan-1-ol is associated to Sercial wines. Verdelho wines are most correlated with 5-(ethoxymethyl)-furfural, nonanone and cis-9-ethyldecenoate. A 96.4% of prediction ability was obtained by the application of stepwise linear discriminant analysis (SLDA) using the 19 variables that maximise the variance of the initial data set.

Content in metallic ions of wines from the Madeira and Azores archipelagos

This study determines for the first time Na, K, Ca, Mg, Fe, Cu, Zn, Mn, Sr, Li and Rb contents in wines from the archipelagos of Madeira and Azores (Portugal). The greater part of the mean content for the different parameters fell within the ranges described in the literature, except for sodium whose higher content may be due to the effect of marine spray. ANOVA was used to establish the metals with significant differences in mean content between the wines from both archipelagos, between table and liquor wines of Madeira, and between wines of Pico and Terceira Islands from the Azores archipelago. Principal component analysis shows differences in the wines according to the wine-making process and/or the equipment employed. Stepwise linear discriminant analysis achieves a good classification and validation of wines according to the archipelago of origin, and the island in the case of Azores wines.

Investigation of urinary volatile organic metabolites as potential cancer biomarkers by solid-phase microextraction in combination with gas chromatography-mass spectrometry

BACKGROUND: Non-invasive diagnostic strategies aimed at identifying biomarkers of cancer are of great interest for early cancer detection. Urine is potentially a rich source of volatile organic metabolites (VOMs) that can be used as potential cancer biomarkers. Our aim was to develop a generally reliable, rapid, sensitive, and robust analytical method for screening large numbers of urine samples, resulting in a broad spectrum of native VOMs, as a tool to evaluate the potential of these metabolites in the early diagnosis of cancer. METHODS: To investigate urinary volatile metabolites as potential cancer biomarkers, urine samples from 33 cancer patients (oncological group: 14 leukaemia, 12 colorectal and 7 lymphoma) and 21 healthy (control group, cancer-free) individuals were qualitatively and quantitatively analysed. Dynamic solid-phase microextraction in headspace mode (dHS-SPME) using a carboxenpolydimethylsiloxane (CAR/PDMS) sorbent in combination with GC-qMS-based metabolomics was applied to isolate and identify the volatile metabolites. This method provides a potential non-invasive method for early cancer diagnosis as a first approach. To fulfil this objective, three important dHS-SPME experimental parameters that influence extraction efficiency (fibre coating, extraction time and temperature of sampling) were optimised using a univariate optimisation design. The highest extraction efficiency was obtained when sampling was performed at 501C for 60min using samples with high ionic strengths (17% sodium chloride, wv 1) and under agitation. RESULTS: A total of 82 volatile metabolites belonging to distinct chemical classes were identified in the control and oncological groups. Benzene derivatives, terpenoids and phenols were the most common classes for the oncological group, whereas ketones and sulphur compounds were the main classes that were isolated from the urine headspace of healthy subjects. The results demonstrate that compound concentrations were dramatically different between cancer patients and healthy volunteers. The positive rates of 16 patients among the 82 identified were found to be statistically different (Po0.05). A significant increase in the peak area of 2-methyl3-phenyl-2-propenal, p-cymene, anisole, 4-methyl-phenol and 1,2-dihydro-1,1,6-trimethyl-naphthalene in cancer patients was observed. On average, statistically significant lower abundances of dimethyl disulphide were found in cancer patients. CONCLUSIONS: Gas chromatographic peak areas were submitted to multivariate analysis (principal component analysis and supervised linear discriminant analysis) to visualise clusters within cases and to detect the volatile metabolites that are able to differentiate cancer patients from healthy individuals. Very good discrimination within cancer groups and between cancer and control groups was achieved.

Dynamic headspace solid-phase microextraction combined with one-dimensional gas chromatography–mass spectrometry as a powerful tool to differentiate banana cultivars based on their volatile metabolite profile

In this study the effect of the cultivar on the volatile profile of five different banana varieties was evaluated and determined by dynamic headspace solid-phase microextraction (dHS-SPME) combined with one-dimensional gas chromatography–mass spectrometry (1D-GC–qMS). This approach allowed the definition of a volatile metabolite profile to each banana variety and can be used as pertinent criteria of differentiation. The investigated banana varieties (Dwarf Cavendish, Prata, Maçã, Ouro and Platano) have certified botanical origin and belong to the Musaceae family, the most common genomic group cultivated in Madeira Island (Portugal). The influence of dHS-SPME experimental factors, namely, fibre coating, extraction time and extraction temperature, on the equilibrium headspace analysis was investigated and optimised using univariate optimisation design. A total of 68 volatile organic metabolites (VOMs) were tentatively identified and used to profile the volatile composition in different banana cultivars, thus emphasising the sensitivity and applicability of SPME for establishment of the volatile metabolomic pattern of plant secondary metabolites. Ethyl esters were found to comprise the largest chemical class accounting 80.9%, 86.5%, 51.2%, 90.1% and 6.1% of total peak area for Dwarf Cavendish, Prata, Ouro, Maçã and Platano volatile fraction, respectively. Gas chromatographic peak areas were submitted to multivariate statistical analysis (principal component and stepwise linear discriminant analysis) in order to visualise clusters within samples and to detect the volatile metabolites able to differentiate banana cultivars. The application of the multivariate analysis on the VOMs data set resulted in predictive abilities of 90% as evaluated by the cross-validation procedure.

Characterization of volatile substances in apples from rosaceae family by headspace solid‐phase microextraction followed by GC‐qMS

The volatile composition of different apple varieties of Malus domestica Borkh. species from different geographic regions at Madeira Islands, namely Ponta do Pargo (PP), Porto Santo (PS), and Santo da Serra (SS) was established by headspace solid-phase microextraction (HS-SPME) procedure followed by GC-MS (GC-qMS) analysis. Significant parameters affecting sorption process such as fiber coating, extraction temperature,extractiontime,sampleamount,dilutionfactor,ionicstrength,anddesorption time,wereoptimizedanddiscussed.TheSPMEfibercoatedwith50/30 lmdivinylbenzene/carboxen/PDMS (DVB/CAR/PDMS) afforded highest extraction efficiency of volatile compounds, providing the best sensitivity for the target volatiles, particularly whenthesampleswereextractedat508Cfor30 minwithconstantmagneticstirring. A qualitative and semi-quantitative analysis between the investigated apple species has been established. It was possible to identify about 100 of volatile compounds amongpulp(46,45,and39),peel(64,60,and64),andentirefruit(65,43,and50)inPP, PS,andSSapples,respectively.Ethylesters,terpenes,andhigheralcoholswerefound tobethemostrepresentativevolatiles. a-Farnesene,hexan-1-olandhexyl2-methylbutyratewerethecompoundsfoundinthevolatileprofileofstudiedappleswiththelargestGCarea,representing,onaverage,24.71,14.06,and10.80%ofthetotalvolatilefractionfromPP,PS,andSSapples.InPPentireapple,themostabundantcompoundsidentified were a-farnesene (30.49%), the unknown compound m/z (69, 101, 157) (21.82%) andhexylacetate(6.57%).RegardingPSentireapplethemajorcompoundswere a-farnesene(16.87%),estragole(15.43%),hexan-1-ol(10.94),andE-2-hexenal(10.67).a-Farnesene(30.3%),hexan-1-ol(18.90%),2-methylbutanoicacid(4.7%),andpentan-1-ol(4.6%) werealsofoundasSSentireapplevolatilespresentinahigherrelativecontent.Principal component analysis (PCA) of the results clustered the apples into three groups according to geographic origin. Linear discriminant analysis (LDA) was performed in order to detect the volatile compounds able to differentiate the three kinds of apples investigated. The most important contributions to the differentiation of the PP, PS, and SS apples were ethyl hexanoate, hexyl 2-methylbutyrate, E,E-2,4-heptadienal, pethylstyrene,andE-2-hexenal.

Previsão do índice bovespa por meio de redes neurais artificiais: uma análise comparada aos métodos tradicionais de séries de tempo

Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model

Comparação de alguns modelos matemáticos para o ajuste às curvas de lactação individuais de vacas da raça Caracu

Foram ajustadas 7239 curvas de lactação de vacas Caracu, controladas semanalmente entre os anos de 1978 a 1988, pertencentes à Fazenda Chiqueirão, Poços de Caldas, MG. As funções utilizadas foram a linear hiperbólica (FLH), a quadrática logarítmica (FQL), a gama incompleta (FGI) e a polinomial inversa (FPI). Os parâmetros foram estimados por meio de regressões não lineares, usando-se processos iterativos. A verificação da qualidade do ajuste baseou-se no coeficiente de determinação ajustado (R²A), no teste de Durbin-Watson (DW) e nas médias e desvios-padrão estimados para os parâmetros e funções dos parâmetros dos modelos. Para a curva média, os R²A foram superiores a 0,90 para todas as funções. Bons ajustes, baseados nos R²A>0,80 foram obtidos, respectivamente, por 25,2%, 39,1%, 31,1% e 28,4% das lactações ajustadas pelas funções FLH, FQL, FGI e FPI. de acordo com o teste de DW, bons ajustes foram proporcionados para 29,4% das lactações ajustadas pela FLH, 54,9% pela FQL, 34,9% pela FGI e 29,6% pela FPI. Para ambos os critérios, a FQL foi superior às demais funções, indicando grande variação nas formas das curvas de lactação geradas pelos ajustes individuais. Curvas atípicas foram estimadas pelas funções, com picos ocorrendo antes do parto e algumas vezes após o término da lactação. Todas as funções apresentaram problemas quando ajustaram dados individuais.

Estudo da água produzida em diferentes zonas de produção de petróleo, utilizando a hidroquímica e a análise estatística de parâmetros químicos

Over exploitation of oil deposits on land onshore or offshore, there is simultaneous generation of waste water, known as produced water, which represents the largest waste stream in the production of crude oil. The relationship between the chemical composition of oil and water production and the conditions in which this process occurs or is favored are still poorly studied. The area chosen for the study has an important oil reserve and an important aquifer saturated with freshwater meteoric. The aim of this work is to study some chemical parameters in water produced for each reservoir zone of production in mature oil fields of Açu Formation, using the hydrochemical and statistical analysis to serve as a reference and be used as tools against the indicator ranges water producers in oil producing wells. Samples were collected from different wells in 6 different areas of production and were measured 50 parameters, which can be classified into three groups: anions, cations and physicochemical properties (considering only the parameters that generated values above detection limits in all samples). Through the characterization hydrochemistry observed an area of water and chlorinated sodium, chlorinated calcium or magnesium (mixed) in well water in different areas of Açu, by applying a statistical treatment, we obtained a discriminant function that distinguishes chemically production areas. Thus, it was possible to calculate the rate of correct classification of the function was 76.3%. To validate this model the accuracy rate was 86%

Feature selection through gravitational search algorithm

In this paper we deal with the problem of feature selection by introducing a new approach based on Gravitational Search Algorithm (GSA). The proposed algorithm combines the optimization behavior of GSA together with the speed of Optimum-Path Forest (OPF) classifier in order to provide a fast and accurate framework for feature selection. Experiments on datasets obtained from a wide range of applications, such as vowel recognition, image classification and fraud detection in power distribution systems are conducted in order to asses the robustness of the proposed technique against Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and a Particle Swarm Optimization (PSO)-based algorithm for feature selection.

Morphological and morphometric comparisons of the first zoea of four species of purse crabs (Decapoda: Brachyura: Leucosiidae) from the Brazilian South Atlantic

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Microhabitat selection and daily movements of two rodents (Necromys lasiurus and Oryzomys scotti) in Brazilian Cerrado, as revealed by a spool-and-line device

Studies on patterns of habitat use by mammals are necessary for understanding the mechanisms involved in their distribution and abundance. In this study, we used the spool-and-line method to investigate habitat utilization by two sigmodontine rodents from Brazilian Cerrado, Necromys lasiurus and Oryzomys scotti. We conducted the study in a Cerrado area in central Brazil (15 degrees 56'S and e 47 degrees 56'W) where the animals were caught in an area of 7.68 ha of Cerrado sensu stricto. Captured individuals were marked, equipped with a spool-and-Line device, and released at the same capture point. The next day we followed the thread to record their daily movements and find their nests. To investigate microhabitat selection we compared habitat characteristics along traits of each studied species with general habitat characteristics of the study area. Although the mean 24-h distance was greater for N. lasiurus (mean +/- SE: 41.9 +/- 42.2 m, N=3) than for O. scotti (28.7 +/- 14.2 m, N=6) this difference was not significant (Mann-Whitney test, U=26, P>0.6). We detected significant differences among observed microhabitats variables of both species and available microhabitat characteristics as determined by discriminant analysis (Wilks's lambda F=3.001; df=14, 116; P<0.001). Both species were associated to microhabitat characteristics whose values differed markedly from the overall available habitat. Along the first canonical discriminant function of the DFA both them were associated with greater grass height than the mean height available and along the second axis N. lasiurus selected areas with higher fruit availability and more shelters than those selected by 0. scotti. For stronger inferences regarding differential patterns of habitat utilization by Cerrado rodents we suggest the simultaneous use of both spool-and-line and standard trapping methods. (c) 2005 Deutsche Geseltschaft fur Saugetierkunde. Published by Elsevier GmbH. ALL rights reserved.

DNA barcode discovers two cryptic species and two geographical radiations in the invasive drosophilid Zaprionus indianus

Comparing introduced to ancestral populations within a phylogeographical context is crucial in any study aiming to understand the ecological genetics of an invasive species. Zaprionus indianus is a cosmopolitan drosophilid that has recently succeeded to expand its geographical range upon three continents (Africa, Asia and the Americas). We studied the distribution of mitochondrial DNA (mtDNA) haplotypes for two genes (CO-I and CO-II) among 23 geographical populations. mtDNA revealed the presence of two well-supported phylogenetic lineages (phylads), with bootstrap value of 100%. Phylad I included three African populations, reinforcing the African-origin hypothesis of the species. Within phylad II, a distinct phylogeographical pattern was discovered: Atlantic populations (from the Americas and Madeira) were closer to the ancestral African populations than to Eastern ones (from Madagascar, Middle East and India). This means that during its passage from endemism to cosmopolitanism, Z. indianus exhibited two independent radiations, the older (the Eastern) to the East, and the younger (the Atlantic) to the West. Discriminant function analysis using 13 morphometrical characters was also able to discriminate between the two molecular phylads (93.34 +/- 1.67%), although detailed morphological analysis of male genitalia using scanning electron microscopy showed no significant differences. Finally, crossing experiments revealed the presence of reproductive barrier between populations from the two phylads, and further between populations within phylad I. Hence, a bona species status was assigned to two new, cryptic species: Zaprionus africanus and Zaprionus gabonicus, and both were encompassed along with Z. indianus and Zaprionus megalorchis into the indianus complex. The ecology of these two species reveals that they are forest dwellers, which explains their restricted endemic distribution, in contrast to their relative cosmopolitan Z. indianus, known to be a human-commensal. Our results reconfirm the great utility of mtDNA at both inter- and intraspecific analyses within the frame of an integrated taxonomical project.

Castes and asynchronous colony cycle in Polybia bistriata (Fabricius) (Hymenoptera: Vespidae)

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)