Comparative analysis between statistical and artificial intelligence models in business failure prediction

A growing number of predicting corporate failure models has emerged since 60s. Economic and social consequences of business failure can be dramatic, thus it is not surprise that the issue has been of growing interest in academic research as well as in business context. The main purpose of this study is to compare the predictive ability of five developed models based on three statistical techniques (Discriminant Analysis, Logit and Probit) and two models based on Artificial Intelligence (Neural Networks and Rough Sets). The five models were employed to a dataset of 420 non-bankrupt firms and 125 bankrupt firms belonging to the textile and clothing industry, over the period 2003–09. Results show that all the models performed well, with an overall correct classification level higher than 90%, and a type II error always less than 2%. The type I error increases as we move away from the year prior to failure. Our models contribute to the discussion of corporate financial distress causes. Moreover it can be used to assist decisions of creditors, investors and auditors. Additionally, this research can be of great contribution to devisers of national economic policies that aim to reduce industrial unemployment.

Comparative Analysis between Statistical and Artificial Intelligence Models in Business Failure Prediction

A growing number of predicting corporate failure models has emerged since 60s. Economic and social consequences of business failure can be dramatic, thus it is not surprise that the issue has been of growing interest in academic research as well as in business context. The main purpose of this study is to compare the predictive ability of five developed models based on three statistical techniques (Discriminant Analysis, Logit and Probit) and two models based on Artificial Intelligence (Neural Networks and Rough Sets). The five models were employed to a dataset of 420 non-bankrupt firms and 125 bankrupt firms belonging to the textile and clothing industry, over the period 2003–09. Results show that all the models performed well, with an overall correct classification level higher than 90%, and a type II error always less than 2%. The type I error increases as we move away from the year prior to failure. Our models contribute to the discussion of corporate financial distress causes. Moreover it can be used to assist decisions of creditors, investors and auditors. Additionally, this research can be of great contribution to devisers of national economic policies that aim to reduce industrial unemployment.

Statistical classification of road pavements using near field vehicle rolling noise measurements

Low noise surfaces have been increasingly considered as a viable and cost-effective alternative to acoustical barriers. However, road planners and administrators frequently lack information on the correlation between the type of road surface and the resulting noise emission profile. To address this problem, a method to identify and classify different types of road pavements was developed, whereby near field road noise is analyzed using statistical learning methods. The vehicle rolling sound signal near the tires and close to the road surface was acquired by two microphones in a special arrangement which implements the Close-Proximity method. A set of features, characterizing the properties of the road pavement, was extracted from the corresponding sound profiles. A feature selection method was used to automatically select those that are most relevant in predicting the type of pavement, while reducing the computational cost. A set of different types of road pavement segments were tested and the performance of the classifier was evaluated. Results of pavement classification performed during a road journey are presented on a map, together with geographical data. This procedure leads to a considerable improvement in the quality of road pavement noise data, thereby increasing the accuracy of road traffic noise prediction models.

Statistical motion learning for improved transform domain Wyner-Ziv video coding

Wyner - Ziv (WZ) video coding is a particular case of distributed video coding (DVC), the recent video coding paradigm based on the Slepian - Wolf and Wyner - Ziv theorems which exploits the source temporal correlation at the decoder and not at the encoder as in predictive video coding. Although some progress has been made in the last years, WZ video coding is still far from the compression performance of predictive video coding, especially for high and complex motion contents. The WZ video codec adopted in this study is based on a transform domain WZ video coding architecture with feedback channel-driven rate control, whose modules have been improved with some recent coding tools. This study proposes a novel motion learning approach to successively improve the rate-distortion (RD) performance of the WZ video codec as the decoding proceeds, making use of the already decoded transform bands to improve the decoding process for the remaining transform bands. The results obtained reveal gains up to 2.3 dB in the RD curves against the performance for the same codec without the proposed motion learning approach for high motion sequences and long group of pictures (GOP) sizes.

Avaliação da influência dos parâmetros de ensaio no índice de fluidez de termoplásticos

O presente trabalho teve como objectivos avaliar a influência de diversas grandezas e parâmetros de ensaio no índice de fluidez de termoplásticos e calcular a incerteza associada às determinações. Numa primeira fase, procedeu-se à identificação dos principais parâmetros que influenciam a determinação do índice de fluidez, tendo sido seleccionados a temperatura do plastómetro, o peso de carga, o diâmetro da fieira, o comprimento da medição, o tipo de corte e o número de provetes. Para avaliar a influência destes parâmetros na medição do índice de fluidez, optou-se pela realização de um planeamento de experiências, o qual foi dividido em três etapas. Para o tratamento dos resultados obtidos utilizou-se como ferramenta a análise de variância. Após a completa análise dos desenhos factoriais, verificou-se que os efeitos dos factores temperatura do plastómetro, peso de carga e diâmetro da fieira apresentam um importante significado estatístico na medição do índice de fluidez. Na segunda fase, procedeu-se ao cálculo da incerteza associada às medições. Para tal seleccionou-se um dos métodos mais usuais, referido no Guia para a Expressão da Incerteza da Medição, conhecido como método GUM, e pela utilização da abordagem “passo a passo”. Inicialmente, foi necessária a construção de um modelo matemático para a medição do índice de fluidez que relacionasse os diferentes parâmetros utilizados. Foi estudado o comportamento de cada um dos parâmetros através da utilização de duas funções, recorrendo-se novamente à análise de variância. Através da lei de propagação das incertezas foi possível determinar a incerteza padrão combinada,e após estimativa do número de graus de liberdade, foi possível determinar o valor do coeficiente de expansão. Finalmente determinou-se a incerteza expandida da medição, relativa à determinação do índice de fluidez em volume.

Methodology in the epidemiological research of respiratory diseases and environmental pollution

There are complex and diverse methodological problems involved in the clinical and epidemiological study of respiratory diseases and their etiological factors. The association of urban growth, industrialization and environmental deterioration with respiratory diseases makes it necessary to pay more attention to this research area with a multidisciplinary approach. Appropriate study designs and statistical techniques to analyze and improve our understanding of the pathological events and their causes must be implemented to reduce the growing morbidity and mortality through better preventive actions and health programs. The objective of the article is to review the most common methodological problems in this research area and to present the most available statistical tools used.

Analysis of tongue shape and motion in speech production using statistical modeling

The mechanisms of speech production are complex and have been raising attention from researchers of both medical and computer vision fields. In the speech production mechanism, the articulator’s study is a complex issue, since they have a high level of freedom along this process, namely the tongue, which instigates a problem in its control and observation. In this work it is automatically characterized the tongues shape during the articulation of the oral vowels of Portuguese European by using statistical modeling on MR-images. A point distribution model is built from a set of images collected during artificially sustained articulations of Portuguese European sounds, which can extract the main characteristics of the motion of the tongue. The model built in this work allows under standing more clearly the dynamic speech events involved during sustained articulations. The tongue shape model built can also be useful for speech rehabilitation purposes, specifically to recognize the compensatory movements of the articulators during speech production.

Statistical analysis of quality control measurements in IMRT for head and neck and prostate cancer

Intensity Modulated Radiotherapy (IMRT) is a technique introduced to shape more precisely the dose distributions to the tumour, providing a higher dose escalation in the volume to irradiate and simultaneously decreasing the dose in the organs at risk which consequently reduces the treatment toxicity. This technique is widely used in prostate and head and neck (H&N) tumours. Given the complexity and the use of high doses in this technique it’s necessary to ensure as a safe and secure administration of the treatment, through the use of quality control programmes for IMRT. The purpose of this study was to evaluate statistically the quality control measurements that are made for the IMRT plans in prostate and H&N patients, before the beginning of the treatment, analysing their variations, the percentage of rejected and repeated measurements, the average, standard deviations and the proportion relations.

A few what-ifs on using statistical analysis of stochastic simulation runs to extract timeliness properties

Modern real-time systems, with a more flexible and adaptive nature, demand approaches for timeliness evaluation based on probabilistic measures of meeting deadlines. In this context, simulation can emerge as an adequate solution to understand and analyze the timing behaviour of actual systems. However, care must be taken with the obtained outputs under the penalty of obtaining results with lack of credibility. Particularly important is to consider that we are more interested in values from the tail of a probability distribution (near worst-case probabilities), instead of deriving confidence on mean values. We approach this subject by considering the random nature of simulation output data. We will start by discussing well known approaches for estimating distributions out of simulation output, and the confidence which can be applied to its mean values. This is the basis for a discussion on the applicability of such approaches to derive confidence on the tail of distributions, where the worst-case is expected to be.

From simulation to statistical analysis: timeliness assessment of ethernet/IP-based distributed systems

A number of characteristics are boosting the eagerness of extending Ethernet to also cover factory-floor distributed real-time applications. Full-duplex links, non-blocking and priority-based switching, bandwidth availability, just to mention a few, are characteristics upon which that eagerness is building up. But, will Ethernet technologies really manage to replace traditional Fieldbus networks? Ethernet technology, by itself, does not include features above the lower layers of the OSI communication model. In the past few years, it is particularly significant the considerable amount of work that has been devoted to the timing analysis of Ethernet-based technologies. It happens, however, that the majority of those works are restricted to the analysis of sub-sets of the overall computing and communication system, thus without addressing timeliness at a holistic level. To this end, we are addressing a few inter-linked research topics with the purpose of setting a framework for the development of tools suitable to extract temporal properties of Commercial-Off-The-Shelf (COTS) Ethernet-based factory-floor distributed systems. This framework is being applied to a specific COTS technology, Ethernet/IP. In this paper, we reason about the modelling and simulation of Ethernet/IP-based systems, and on the use of statistical analysis techniques to provide usable results. Discrete event simulation models of a distributed system can be a powerful tool for the timeliness evaluation of the overall system, but particular care must be taken with the results provided by traditional statistical analysis techniques.

Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

Assessment of climate change statistical downscaling methods: Application and comparison of two statistical methods to a single site in Lisbon

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia do Ambiente

Algebraic structure for the crossing of balanced and stair nested designs

Stair nesting allows us to work with fewer observations than the most usual form of nesting, the balanced nesting. In the case of stair nesting the amount of information for the different factors is more evenly distributed. This new design leads to greater economy, because we can work with fewer observations. In this work we present the algebraic structure of the cross of balanced nested and stair nested designs, using binary operations on commutative Jordan algebras. This new cross requires fewer observations than the usual cross balanced nested designs and it is easy to carry out inference.